- Memo to Microsoft: The readers respond
- Halloween Fun: iPod Linux Frankenstein
- Oracle doesn't get it: open source is not a price tag
- Levanta releases MapFS under GPL
- Open source and The Keystone Advantage
- JBoss is not impressed with IBM's offering
- Microsoft continues to grow
- Write code, get $50K and a free trip to OSBC
- New Web 2.0 company: TagTagger
- Memo to Microsoft : Buy Novell or JBoss, or both
October 31, 2005 | Comments: (0)
Memo to Microsoft: The readers respond
The comments related to Memo to Microsoft: Buy Novell or JBoss, or both have been pretty interesting. Some basically accuse me of spreading MS FUD, which I assure you I am not inclined to do, and others suggest that the anti-trust impact of MS buying Novell would be detrimental to Microsofts' very existence (I disagree.) The argument that rings most relevant to me is the impact open source licenses like the GPL would have on an acquisition-and this goes for any company, not just MS. It made me realize how important license choice is, and makes wonder how much it affects the valuation of open source companies in general. But a license doesn't necessarily preclude forking GPL'd code into another OS. Regardless, MS could absorb Novell or even RedHat and keep the distros open source without a major financial impact-licensing revenue would just be an additional revenue stream.
We tend to think that MS is either full of FUD or running scared of Linux. The fact of the matter is that they could easily swoop in and have a very big impact on the market. The goal of marketing is to create a monopoly. If you don't think that Microsoft could create another monopoly you are sadly mistaken.
The whole point of the post was to try to get people to think about the technology market. Open source hasn't had a big moment like Oracle's purchase of PeopleSoft-the closest thing was Oracle's purchase of InnoDB. Those of us involved in open source tend to get a little wrapped up in "which is the best distro" and don't spend enough time thinking about the impact on the broader landscape.
Posted by Dave Rosenberg on October 31, 2005 10:56 AM
October 31, 2005 | Comments: (0)
Halloween Fun: iPod Linux Frankenstein
All this talk about open source is meaningless if people aren't actually doing something with the code. In this case, that something is making Linux run on an iPod. It even has a GUI called FloydZilla. Who said open source didn't innovate?
Posted by Dave Rosenberg on October 31, 2005 08:55 AM
October 31, 2005 | Comments: (0)
Oracle doesn't get it: open source is not a price tag
ZDNet is reporting that Oracle is due to announce a stripped-down, free database. Why? To try to defend against the rising tide of open source databases:
The database heavyweight on Tuesday is expected to announce the beta release of Oracle 10g Express Edition (Oracle Database XE), which will be generally available by the end of the year. It is targeted at students, small organizations and software vendors that could embed the Oracle database with an application.In other words, it mostly stinks, but at least it's free! What Oracle apparently hasn't understood is that free, as in price, is just one part of the open source puzzle. But it's not necessarily the most important one.The latest edition is the same as other databases in Oracle's lineup but is limited in usage. It can only run servers with one processor, with 4GB of disk memory and 1GB of memory.
My prediction? This move will be completely forgotten. Few to nobody will use it. And, 6-12 months from now, Oracle will have to give a real response to the open source threat it faces. Tossing a lightweight database in front of a fast-moving market that wants free, open, and killer databases just won't fly.
You can do better than this, Oracle.
Posted by Matt Asay on October 31, 2005 05:18 AM
October 30, 2005 | Comments: (0)
Levanta releases MapFS under GPL
Linux management vendor Levanta released MapFS under the GPL. MapFS is a virtualized Linux kernel-loadable module that provides virtualization of various Linux filesystems. It makes Linux systems easier to manage because it allows the files of multiple Linux boxes spread throughout a datacenter to exist in a central location within very little disk space.
MapFS implements a Linux filesystem which utilizes copy-on-write functionality and existing Linux filesystems to allow component filesystems (or portions thereof) to be combined into a single virtual filesystem that appears to be fully writable. MapFS is written in C and uses the standard Linux kernel VFS and loadable module interfaces for defining new filesystem types to the kernel. MapFS supports major kernel versions, including 2.4.7 > 2.6.13.This functionality significantly eases data sharing between multiple machines connected to a shared storage medium (SAN/NAS/Mainframe DASD) as data can be optimistically shared between all of the machines in a way which is completely transparent to the applications running on them. MapFS will execute a copy-on-write to obtain a private copy in the event that one of the Machines attempts to modify a copy that is on the shared medium.
Savvy Linux file or kernel gurus will be reminded of UnionFS, but there are some fundamental differences. At a high level, UnionFS provides for disperse system directories to be unified, while MapFS provides individual files to be enumerated and acted-upon via pointer-like operations, completely independent of the directories from which they came. Think of it as providing the next level of virtualization and abstraction for any Linux file, regardless of where it came from, or where it needs to be accessed. This translates into a entire Linux distribution using MapFS, and all of its data, being extremely portable, space-efficient, and quick to change.
With the release of MapFS, Levanta proves that it's not just paying lip service to open source-it's donating developer time and useful code. Other companies should take note and follow the example. As Matt noted in Open source and The Keystone Advantage, the best companies create as much opportunity as they consume.
Posted by Dave Rosenberg on October 30, 2005 09:00 PM
October 28, 2005 | Comments: (0)
Open source and The Keystone Advantage
I just finished plowing through The Keystone Advantage, a very good book that Arjen recommended to me. (Many thanks, Arjen!)
Here's the premise. We live in an interconnected world, but sometimes in technology we don't fully embrace this concept. Or, if we do, we don't behave in ways that capitalize on the insight and help our firms. The technology landscape is a network of networks, with keystones at the hubs, niche players surrounding the hubs, and dominators consuming the hubs and niche players. Understanding one's role in the technology universe (most of us are niche players), and playing that role well, can help us to thrive as a company.
Simple premise. But the book offers some insights that I found valuable. But first, a bit more detail on the terminology:
Keystone and dominator strategies can be pursued by firms that occupy important hubs in their business networks. Keystone strategies shape and coordinate the ecosystem, largely by the dissemination of platforms that form a foundation for ecosystem innovation and operations. These contrast with dominator strategies, which attack the ecosystem, absorbing and integrating external assets into internal operations. Niche strategies can be pursued by the much larger number of firms that make up the bulk of the ecosystem. These firms emphasize differentiation by focusing on unique capabilities and leveraging key assets provided by others. (10)With this last point in mind, the authors suggest that "the crucial battle is not between individual firms but between networks of firms. Innovation and operations have become a collective activity" (11).
Exactly. This, by the way, is how we need to think about open source. An ecosystem. A network. "The goal is no longer to lock out entire vertical stacks with proprietary advantage, but to be the best in a chosen area of specialization" (23). Bingo. Let Alfresco be the best in content management, SugarCRM in CRM, JBoss in application servers, etc. Tie them together (maybe with SpikeSource?), each focusing on what it does best, and use that to deliver value to customers. The sum much greater than the individual parts.
Some excerpts with commentary:
Both Wal-Mart and Microsoft deliberately chose to shape the collective performance of the networks of firms that depend on them by offering platforms on which others could build. Both firms were successful because they appreciated the impact these platforms could have on these business networks and took steps to realize this impact by creating real opportunities for other firms. (3)Iansiti and Levien (the authors) make this important point throughout the book: the best companies create as much opportunity as they consume. Dominators suck value out of a network, whereas keystones enable and foster value. This isn't to say that they don't often compete with their niche players (the constellation of partners/competitors/etc. that live in the ecosystem enabled by their platform), but that the overall effect they have on an industry is positive.
I believe Microsoft was definitely a "keystone" in its youth (and much more open than its mainframe peers (31), but has been forced to consume the value networks around it in order to grow. At its most basic level, this can be seen in the API counts in Windows (no longer the baseline operating system it once was, Microsoft continues to dump everything the US Justice Department will allow it to put into Windows), as the authors discuss (56), but a more detailed view emerges once you look at its acquisitions in the past few years. It's clearly on a "domination" path.
The authors disagree (see especially 85), but I don't think they fully appreciate what Microsoft's acquisitions of Great Plains and other companies mean for the industry. Microsoft's "integrated innovation" translates into one massive software offering - an ecosystem/network in and of itself. I don't see how this can fit the authors' definition of a keystone. It sounds precisely like a dominator.
What we need, however, are more keystones. Or, rather, better keystones. I think Novell or Red Hat make ideal keystones, but Red Hat doesn't seem interested in an open source ecosystem (my subjective, biased view, but one that seems justified by the many open source startups that can't seem to catch Red Hat's attention, despite doing very interesting things) while Novell is still in the early stages of growing one (its MarketStart program is an example).
Hubs can play an enormously important role in the collective performance of a network. They have the potential to increase the ease with which different network nodes connect to each other and thereby decrease the complexity involved in the coordination and integration efforts necessary for improving productivity and achieving growth....Again, keystones thrive because they help others to thrive. It's not about sucking value out of a network. It's about putting in as much as you get out, and making sure your partners (and, in a way, your competitors) also thrive. There's much more money in a healthy ecosystem than in a monopoly.The biological literature suggests that a species that serves as a hub in a food web or other network of ecosystem interactions can improve its overall chances of survival in the face of change by providing benfits to the ecosystem as a whole....Removal of biological keystones can have dramatic cascading effects through the entire ecosystem , whereas removal of other species, even species involved in many interactions, can have little effect beyond the loss of those connections....
Keystones can enhance the productivity of their ecosystems in a variety of ways. Some do so by directly removing or limiting the number of species that would otherwise disproportionately reduce productivity. [A possible counter to my statement about Red Hat above? i.e., Maybe Red Hat plays a filtering role?]....Keystones can also enhance productivity from the bottom up by providing a foundation on which other species rely....
Naturally, the most direct way for a keystone to ensure its continued survival is to directly maintain the stability of its ecosystem. Therefore it is to be expected that keystones act to encourage stability directly in addition to encouraging it indirectly through their effect on diversity. (67-70)
On the latter...
A classic dominator acts to integrate vertically or horizontally to directly control and own a large proportion of a network. In so doing, dominators become directly and solely responsible for the majority of both value capture and value creation in their networks, leaving little opportunity for the emergence of a meaningful ecosystem. A value dominator, or hub landlord, in contrast, eschews control of the network and instead pursues control of value extraction alone [like Enron]. In so doing it provides little new value to its network, while at the same time taking what value there is for itself, leaving a starved and unstable ecosystem around it. (74)In either scenario, the dominator is a net consumer of value in the ecosystem, at least as others are concerned.
It's an interesting read, with direct implications on the budding open source ecosystem. I don't think we have any strong keystones yet, though each member of the LAMP stack (+ JBoss) has the ability to become such. MySQL strikes me as being well on its way. I also think that some players can be both niche players in the wider industry, and keystones within their subsegments. So, SugarCRM is creating an open source CRM "market" (network) around itself, as are Alfresco and others.
Open source is a community of communities of communities. Those that recognize this and actively try to create and share value will be the "keystones" around which this industry revives and thrives
Posted by Matt Asay on October 28, 2005 06:20 PM
October 28, 2005 | Comments: (0)
JBoss is not impressed with IBM's offering
Gavin King of Hibernate/JBoss fame dropped us a comment regarding the impact of IBM's WebSphere announcement on JBoss.
While Derby is an interesting product, and while I very much hope that the pure-Java database space matures soon, I don't think it's at all reasonable to believe that people will be deploying production web or enterprise applications on Derby in the near future. (This is certainly a niche for pure Java databases, but this is probably not it.) Rather, people in pursuit of a complete-opensource stack will continue to use MySQL or PostgreSQL as they do today. We have a great relationship with MySQL, and I don't frankly see Derby as any kind of a threat to their market penetration. The JBoss / MySQL stack is a many times more credible combination than WebSphere Community / Derby! Meanwhile, our customers have huge existing legacy databases and most aren't interested in moving off of Oracle or DB2 or Sybase or SQL Server or Teradata or whatever at the same time they migrate away from WebSphere or WebLogic. I mean, why should they? In most cases, switching databases is going to be a much, much more expensive proposition than switching appservers.So, in conclusion, I don't see there being any real synergy in having a database bundled with the application server. Frankly, if I were IBM, I would not be trumpeting my story around data without having some (any) kind of credible story for /persistence/ first. Hehe, I should say that the proposed stacks are:
WebSphere Community / EJB CMP / Derby
JBoss AS / Hibernate / MySQL
Gavin makes a good point. I don't think that Derby is really all that relevant. The issue that remains important is that IBM is capable of creating a good stack, and with it's sales force could definitely sway customers to purchase the whole package from them. The best product doesn't always win-if it did we'd all be using Macs ;>
BTW-Thanks to Gavin for shutting me up.
Posted by Dave Rosenberg on October 28, 2005 12:13 PM
October 28, 2005 | Comments: (0)
Say what you will, but Microsoft is not showing signs of going away. Yes, its MSN unit has been weak relative to Yahoo! and Google, but MSN has always been a bit of a dog. (Sorry, Jason.) The Xbox is also apparently not doing as well as expected, though from my perspective (I have a few teenage neighbors who seem to do nothing besides play Xbox all day and night) it's thriving to the detriment of humankind. :-) Windows Mobile, however, is on a 50% boom.
Again, for a company Microsoft's size, a 6% rise in revenues and 16% rise in operating profit is exceptional. Microsoft has its problems, but making money is not yet one of them. Looks like I need to step up my efforts to bury it.... :-)
Posted by Matt Asay on October 28, 2005 08:55 AM
October 28, 2005 | Comments: (0)
Write code, get $50K and a free trip to OSBC
This popped up on my OSBC Google alert this morning. (Yes, I criticize them, but I still use them. :-) I had no idea, but am flattered that OSBC has become a destination resort of sorts. Here's the scoop:
Realm Systems is giving away $50,000 and a free trip to the Open Source Business Conference in February (14-15, 2006). What do you have to do to get it? (And which is the bigger prize, the money or the insight you'll glean at OSBC? OK, I agree. It's the money.)
From the press release:
Realm Systems announces the start of Project BlackDog, a skills contest offering prizes in five categories, including a $50,000 grand prize for the best application created or ported to run on BlackDog(TM). The contest started October 15 and ends January 15, 2006. Prizes will also be awarded for the most bugs reported that are determined to be critical to BlackDog functionality. Hundreds of developers have already received their units in anticipation of creating the grand-prize winning application and being recognized at a key industry conference [That would be OSBC].
For those of you unfamiliar with Realm Systems (and BlackDog), Realm is a very cool technology company that has a mobile enterprise platform. What does this mean? I don't know. What it meant the last time I was debriefed by them was that they had created a super-cool mobile personal server that effectively allows you to run your apps (Windows-based, Linux-based, MacOS-based) on any other system. Just plug the little thumb-drive-like device to the USB port and away it goes - no performance degradation, and highly secure. Very cool.
Realm has since expanded its product line, and is looking for help from the open source community, bounty-style. It sounds like Project BlackDog relates to this MPS product:
BlackDog represents a new breed of device that redefines mobile computing. BlackDog is a pocket-sized, fully self-contained Linux server with a built-in biometric reader and Debian-based Linux operating system. Unlike any other mobile computing device, BlackDog contains its own processor, memory and storage, and is completely powered by the USB port of a host computer -- no external power adapter required.It also sounds like the company has scaled back its ambition (slightly) to only allow Linux-based apps to run on other operating systems. Still, that's pretty cool (no WINE required). Take a look, earn some money, and join us at OSBC San Francisco. (Along with Mitch Kapor, Peter Thiel (Founder, PayPal), Kim Polese, Nick Carr (Does IT Matter?), John Roberts (SugarCRM), John Powell (Alfresco), Marten Mickos, and others.)
Posted by Matt Asay on October 28, 2005 08:45 AM
October 28, 2005 | Comments: (0)
New Web 2.0 company: TagTagger
Matthew Langham has done it again. Not content with his day job, Matthew (aka "Silent Penguin") has launched TagTagger, the newest sensation in Web 2.0. I'm a bit giddy just thinking about the possibilities. Tags upon tags upon...tags.
Head over to TagTagger and join the revolution! (And don't miss out on TagCamp. See the site for details, but you definitely want to be on the invite list for this one!)
(Awesome, Matthew. 6:27 AM, and I'm going to wake my kids with my laughing.)
Posted by Matt Asay on October 28, 2005 05:24 AM
October 27, 2005 | Comments: (0)
Memo to Microsoft : Buy Novell or JBoss, or both
In a previous post I suggested that Microsoft buy JBoss. After much thought and discussion, I think that Microsoft should also buy Novell (market cap of $2.8 billion). Not only would this give MS an instant place in the open source market but it would also allow them to cut marketing expenses geared toward killing Linux. With just RedHat left as the main enterprise Linux vendor IBM or HP would have to buy RedHat before MS stomped them out of existence. Novell sits with $1.6 billion in cash and revenues of $1.8 billion. This is far better than RedHat's $517 million in cash and revenues of $235 million, despite what Matt Szulik might lead us to believe.
Or, how about Microsoft buying JBoss? JBoss is a much easier pill to swallow-figure that acquiring JBoss wouldn't cost more than $200 million (being very generous) and MS has $37.75 billion in the bank. I bet they wouldn't even need an executive signature on that low of an expenditure. Think of the fact that Microsoft could own the most prevalent Java app server along with a Linux vendor. It's not that hard to imagine. And we certainly should be afraid.
Alternatively, Sun could buy JBoss and actually have a viable Java server product. The angle that puzzles me is why Sun didn't try to buy Novell or RedHat instead of piddling away $4.1 billion on StorageTek. Maybe they did and we don't know, but my guess is that short-term gain overpowered strategy. If Jonathan Schwartz wants us to believe him when talking about being open, blah, blah, blah where is the proof? CDDL? Nope.
Meanwhile, Sun investors approved the "anti-poison pill" initiative by an overwhelming 84 percent, but a second shareholder proposal to rein-in executive options failed to pass. This means that Sun can now be acquired, but not before Schwartz and McNealy manage to get as much stock as possible. Either way, that would be a shocker, I don't see the corporate culture surviving without McNealy.
How much is Novell actually worth?
Enterprise value (EV) represents a company's economic value -- the minimum someone would have to pay to buy it outright.
To calculate enterprise value, start with a company's market cap, add debt (found on a company's balance sheet), and subtract cash and investments (also on the balance sheet). To get total debt, add together long- and short-term debt.
Novell EV=$2.8b-$1.6b+600m
Novell Enterprise Value=$1.8 billion
Even if the purchase price was double the enterprise value, Novell is still a pretty good buy. Novell sits with $1.6 billion in cash and revenues of $1.8 billion. This is far better than RedHat's $517 million in cash and revenues of $235 million
Sun
Market cap: $13.3 billion
Cash: $3.40 billion
Microsoft
Market cap: $266.21 billion
Cash: $37.75 billion
Novell
Market cap: $2.8 billion
Cash: $1.6 billion
Revenue: $1.8 billion
RedHat
Market cap: $3.8 billion
Cash: $517 million
Revenue: $235 million
JBoss
Funding: $10 million
Valuation (this is a guess): $50 million
Posted by Dave Rosenberg on October 27, 2005 03:39 PM
October 27, 2005 | Comments: (0)
Will the "pretenders" please stand up?
UPDATED
Matthew Szulik has a point. Several, actually. I'm just not sure if any of them are valid. Matthew raises a useful issue, but does so in a way that I find counter-productive.
From the Register:
Szulik...[said] it was wrong to think companies like Red Hat could control what the open source community builds and that it's important to stay true to the premise of the Gnu General Public License (GPL).Agreed. But it does make one wonder why Red Hat employs four of the top 11 Linux kernel developers. [Update: Dave Jones seriously questions this number, as well as its importance. Thanks for the clarification.) Why? Surely there's a reason for this. In part, Red Hat's argument to customers (which I've heard back from them) is that this gives Red Hat superior leverage to support the Linux kernel. Why? Because they have a better sense for where it's going....Ah. Caught. I can think of a few others, but I'd be interested to hear Matthew's rationale (as he once told me that Red Hat didn't buy a certain open source company - though they had actively considered it - because they had access to the source code, anyway, so why bother owning the company when they could access the code? Indeed).
Companies that don't remain true to the GPL or who don't endorse patent-free software violate the concept of open source and are hurting innovation, Szulik said.Like IBM, the company that has more patents than any other, and yet did more to make Linux a commercial reality (sorry, Matthew) than any other company, by a huge margin?
And I'm still waiting to hear why adherence to the GPL (instead of BSD or other open source licenses) is the true mark of an open source purist. I think the Apache guys might have something to say about this....Besides, my recent post notwithstanding, what innovation isn't happening that Matthew is concerned about?
Companies who violate open source, such as those who claim to provide open source but who add "proprietary" layers to the technology, lack legitimacy.Well, I think I know who he's aiming this at, but I also think it falls flat. Not the least reason is that Red Hat has been accused of this very same thing: among others were the comments leveled by r0ml at OSCON. See also Ian's recap.)
Matthew might have found it interesting to watch Richard Stallman castigate Michael Tiemann at the Harvard Business School/MIT Open Source Symposium two years ago - Stallman accused Red Hat of being untrue to open source, flouting the GPL, and generally being a poor open source citizen. Michael, normally erudite, flailed to respond.
But he couldn't. Not because he's not razor-sharp smart (with a wicked-sharp tongue to boot). He is. But it's hard to respond to this sort of criticism, as thoughtless and generic as it is.
Matthew seems to label every open source company besides Red Hat "mudblood." Stallman calls Red Hat "mudblood." The mudbloods call Matthew "self-righteous." And we're no better off than when we started, and much worse. Worse because the name-calling does nothing to help customers,
There are all sorts of flavors of open source, Matthew. There are more business models than the Red Hat model. Please don't look for easy political points by misconstruing others' business models, or by mischaracterizing the effects promising new open source companies are having on the industry, because it ends up hurting all open source companies, and not just the one or two you may have been targeting.
In talking with one of your board members it doesn't seem that RedHat believes there is room for more than 1-3 successful open source companies in the industry (of the kind you describe - the purebloods, not like the mudbloods the rest must be).
I think he's wrong. And I disagree with the tone and substance of your comments at Vortex. I think you'll be happier to be wrong in a world with a thriving open source ecosystem than in a world where Red Hat is king of a very small, exclusive (purebloods only!) open source hill. There's more money in a big industry than by dominating a small one. And no one wins when we degenerate to name-calling and touting who is the most open. The real question is, what will solve customers' business problems. You're doing a good job of that. Be content.
Posted by Matt Asay on October 27, 2005 03:37 PM
October 27, 2005 | Comments: (0)
How much did Google pay for this headline?
UPDATED: Removed some of the language that made it appear that I was criticizing Brian Proffitt (Managing Editor, Linux Today) and his coverage. I wasn't, or didn't intend to.]
We, collectively, seem to have to be on Google meth. How else to explain this headline?
Wow! They must have contributed significant code (and a slew of developers to maintain it and grow a community around it), I thought. After all, that's what I consider model open source citizenship.
But no. They gave money. Not an insignificant amount ($350K to a joint open source technology initiative of Oregon State University and Portland State University) but, again, not what I'd call "model OSS citizenship."
Why? Because money isn't everything. As with charity in "the real world," it's much easier to give money than one's time and expertise. I'd much rather give 10% to my church than 10% of my time. (Unfortunately, if you're LDS, you're generally asked to do both. :-) Money is easy. Time and attention are not.
This isn't meant to downplay Google's commitment of cash (or to criticize Brian for covering a valid, newsworthy story in a way favorable to Google), but rather to suggest it is a fraction of what Google should be giving (yes, I'm waxing normative now). Google, as much or more than most companies on the market right now, is a huge beneficiary of open source. It runs on Linux (which it didn't pay Red Hat much of anything to use - I think they bought one copy of Professional...sorry, Matthew. That's what "real open source companies" can do. However, as nearly as I could tell, Matthew believes only Red Hat is truly an open source company. :-)), and uses a range of other open source software.
Maybe our generosity should be measured not in what we give, but rather in what we keep for ourselves? Or perhaps this is too tough a metric to be helpful. At any rate, I'd like to see Google contributing people to the open source community, and not just cash or code, though the latter would be a big improvement. (To be fair, the company already does contribute people and code. But like Oliver Twist, I wonder: "Please Sir, can we have more?" :-)
Good work, Chris and team. It's a good start. But with what the open source community has gifted you, I think it's fair to ask that Google give more. Fair is fair, after all. What's next?
Posted by Matt Asay on October 27, 2005 03:23 PM
October 26, 2005 | Comments: (0)
IBM launched WebSphere Application Server Community Edition which bundles WebSphere with Derby, and rolls in some technology from Gluecode. This includes a web-based management console and messaging via ActiveMQ (and what do you want to bet there is an MQSeries upgrade available?). The app server is Geronimo, which IBM ingested earlier this year.
This is bad news for JBoss -- they don't have a good database story, and they're battling a much richer product portfolio than they've ever confronted before. Worst of all, IBM's selling services for $900 a server a year. JBoss' price for a one-year subscription STARTS at $3,750.
IBM is the real brand monster here; all along, the risk was that they would see this as an attractive market. It looks as if they have. IBM has a very diverse business and can capitalize a big installed base in ways that a specialist like JBoss never can. Cross-sell, consulting, services, hardware sales and so on are all available to the guys in Armonk, but not the guys in Atlanta.
JBoss has been furiously building out the product line, going mostly up the stack. They're rolling out business process management systems and other distributed apps. The theory -- sound -- is that the app server is about to get commoditized. Witness the free offering from IBM.
So far, though, all JBoss' revenue and virtually all of its growth is from its support sales and subscription offering, and the price of those just fell by seventy-five percent. JBoss hasn't had time to build any traction in the market on these higher-priced and more valuable products. Frankly, that's because most people choose JBoss because it's free, and those guys automatically qualify themselves out of paying big dollars for anything.
Sure, high-end customers will pay for those up-the-stack apps. The question is, can JBoss really compete on business software and applications with IBM?
Unfortunately, I don't see an obvious way for JBoss to compete. Just by announcing this I think IBM forces JBoss to look for an acquirer, rather than to plan for an IPO. I go back to my contention that Microsoft should buy JBoss.
Light your flames...
We've become bored with anonymous comments, we use our real names, the least you could do is use yours.
And before the JBoss PR folk start harassing me (again) keep in mind that the only thing I care about from you are metrics cited that are not accurate. If you want to voice your opinion on why JBoss is great do so in the comments and be sure to let us know what agency you are with. Don't waste your time reading blogs. Go out and get some press.
Posted by Dave Rosenberg on October 26, 2005 07:41 PM
October 26, 2005 | Comments: (0)
Russell Shaw of ZDNet attended the Internet Telephony Conference and Expo yesterday, where he heard Skype CEO speak on Skype's future plans. Some interesting, some not so interesting. (Actually, the most interesting thing to me is Groups, which indicates different Skype demographics than I'd supposed.) But the good news is good enough to finally make the acquisition slightly more intuitive. Here they are. (Thanks, Russell.)
Good stuff. Let's hope that one thing that comes from the eBay acquisition of Skype is that Niklas is actually able to travel to the US. His remote keynotes are getting old. (I've done one before via iChat - it's just a lame way to present.)
- Skype is aggressively planning to open its payment system for ringtones and third-party software. Yes, fellow eBay company PayPal will be a preferred option, but Skype will take credit cards, too. [This is important, in my opinion. Lots of room for widely dispersed payment systems.]
- Skype will be pushing recording systems for Podcasts- which, of course, they would love to enable. [Dear me, are we still talking about podcasts??]
- Voice content enablement, based on subscription models that could be handled via PayPal and other Skype-approved payment methods, will be increasingly marketed. [Not sure I understand precisely what they mean by "voice content," but....]
- More than 1,000 software and hardware developers are affiliated with Skype, with a stream of others on the way.
- Although Skype Groups is intended for business users, Skype is not planning a staffed sales effort to make inroads into enterprise IP telephony. Instead, they will do this via partner products likely to appeal to business users. Zennstrom cited one such example as Skype partner Salesforce.com's integration of Skype links into their CRM system. [This is the biggest news of all, in my book.]
Posted by Matt Asay on October 26, 2005 07:39 AM
October 26, 2005 | Comments: (0)
Research: Open source not innovative? Or simply not loud enough about it?
Krzysztof Klincewicz (Tokyo Institute of Technology; School of Management, Warsaw University) recently published [PDF download] a great first attempt at measuring the levels of innovation in open source projects. "First attempt" because, as Dr. Klincewicz finds, it's very difficult to measure what is, and what isn't, innovative, especially when so much value is attached to that word (and, hence, it is used far more than it should be).
His findings? Using a base of the 500 most actively developed SourceForge projects as his sample (or 3% of overall projects - 81%, or 89,557 registered projects are not active), Klincewicz finds that open source is not very innovative (Only 64 projects, or 12.8% of the sample, were not "direct imitations of existing solutions" (18), and "only 5 out of 500 SourceForge projects could be classified as technological breakthroughs" (20)), but a significant reason for this is the lack of "marketing" within the open source development community. That is, most open source projects (innovative and otherwise) die on the SourceForge "vine" for lack of recognition and outside contribution.
Also, and almost shockingly, 100% of the "radically innovative" projects were started from 1999 - 2001: not a single one has started since then. I guess I have to take back some of my sniping at the Valley. I'm assuming some of the boom-town euphoria of the bubble is what yielded the innovations that Klincewicz finds....
Curiouser and curiouser, to quote Alice. But it gets even more interesting....
Additional findings:
"...[M]any of [the open source community's] development efforts are focused on improving and complementing the Linux platform. Even though certain applications may exist for Windows users, they need to be "re-invented" in the other operating system environment. These re-inventions are not radical technological breakthroughs, as similar benefits and functionality are already available for alternative platforms." (7-8)MNA: It turns out that most open source projects either do the same "new" function, or they're merely re-inventions of old technology on a new platform. Innovative in a way? Sure. But not the kind of innovation most of us associate with the denotation of the word. Here's the specific data:
Why do we see this? Are open source developers simply mindless sheep? I don't think so, and neither does Klincewicz. He laments that the high percentage of "me-too" functionality "creates unnecessary competition for scarce resources within the OSS developer community" (19). But he also finds that the highly formalized decision-making processes within open source communities discourages new concepts, as "[i]t is easier to fork the code...than to convince project decision makers to implement certain ideas" (19).
Strikingly, despite this apparent dearth of innovation (at least, on SourceForge - keep in mind that many of the most popular projects migrate from SourceForge to their own hosted forges), "Innovative projects are significantly more popular among developers than "me-too" solutions" (21). (However, given his own data, I'm not sure how he arrives at this conclusion. His conclusion only holds if you add in the SourceForge project itself, which receives a hugely disproportionate number of feature requests, relative to the other innovative projects he identifies - see page 21).
But if you accept his conclusion, it's clear that developers prefer innovation (why work on the mundane?), but it's simply hard to find it on SourceForge (due to the lack of marketing/discovery resources he identifies and which I mentioned above) (See page 25).
Looking deeper into the identified innovative projects, Klincewicz finds that
"40% of breakthroughs came from company-initiated projects, and 50% of technology modifications grew out of academic research, while community-driven initiatives were in turn more focused on platform modifications and marketing innovations" (22).In other words, much of the new, innovative functionality on SourceForge comes as "feature gifts" from those (companies and academics) outside the open source development community, as traditionally described/conceived. (22)
Time to wring our hands and give up? No. Instead, I think it's time to redefine the open source community, as well as to be a bit more pragmatic about what open source does well, and what it doesn't. For one thing, as Klincewicz notes, all of this bottoms-up commodification has a significant, positive benefit on innovation generally: it forces proprietary firms to race to innovate ahead of the commodification curve (as Larry Augustin detailed [PDF] at the first Open Source Business Conference). Or, as Klincewicz notes, "one could even suggest...that these powerful commoditization mechanisms stimulate more innovation in proprietary than in open software domains" (25; see also Joel West's work).
Assuming this is all true, it means that the most successful open source companies will be those who can create communities to match their projects. In the past, it has been sufficient to find a good project (e.g., Linux), build a support company around it, and fight like mad for branding. This is, in effect, the Red Hat story.
But much of the new functionality, the new open source applications (ECM, ERP, CRM, etc.) will not come from the community, as traditionally defined. It will, almost of necessity, come from companies. Why necessity? Because companies, fueled by VC dollars, can afford the project promotion that Klincewicz finds lacking in the existing open source development community. Again, it's not a question of whether open source can innovate. It's more a question of how to help others hear about it, such that a vibrant community can grow up around it. This is where money is helpful.
Here's to the future community builders, corporate and otherwise.
(And here's a look at some additional interesting data from Klincewicz's research):
Posted by Matt Asay on October 26, 2005 06:26 AM
October 25, 2005 | Comments: (0)
Sun's attempt at a true utility computing model appears to be struggling out of the gates. According to The Register (read at your own risk) Sun has confirmed that they are having a hard time getting customers in general, and failing at getting the few they have to go on record in marketing programs.
"It has been harder than we anticipated," said Aisling MacRunnels, Sun's senior director of utility computing in an interview. "It has been really hard. All of this has been a massive learning experience for us a company. I am not embarrassed to say this because we have been on the leading edge."
I am a big believer in Grid, and the outsourced utility model, so I do find this disappointing. I think that while the market circumstances may indicate that the world may not be ready for the utility model, Sun itself is largely to blame for the lack of customers. Sun's utility model requires many elements to come together that are far more complicated than Grid computing alone-which is already hard. For most organizations, even large scale enterprises, it's just not practical to develop applications to run on an outside system. There are too many variables and ROI questions regarding the development expense. Add to that the fact that Sun has been Sybil-esque in its focus and the equation simply doesn't add up.
I do appreciate Sun's vision for utility computing, and in fact I subscribe to it. I just won't pay Sun for the subscription.
Posted by Dave Rosenberg on October 25, 2005 01:43 PM
October 25, 2005 | Comments: (0)
Self-Indulgence and Silicon Valley
I downloaded Flock today. I was curious to see why it had managed to generate a fair amount of buzz, given all the other hollow buzz around Web 2.0. Maybe this application was different?
Well, it is different, in the sense that it is actually a full-blown application, whereas much of Web 2.0 consists of simple features that can never stand alone as companies. But the fact that Flock is an application is also one of its downfalls - it's a browser. Who needs another browser?
More pertinently to this discussion, who beyond Silicon Valley could have thought that the world needed this kind of browser? It's a browser that aggregates many of Web 2.0's worst buzzwords (RSS feeds! del.ici.ous tags! Flickr photos!), buzzwords that virtually no one outside of Silicon Valley cares about. (It's a bit like a flocked Christmas tree. You know, that white, superficial, ugly stuff that gets sprayed onto Christmas trees to look like snow? Except that it doesn't look like snow - it looks like flock, whatever that is.)
Living in Silicon Valley, it's easy to forget this simple fact: no one else cares about technology as much as the Valley's denizens do. No one. Other people make money in other ways, and so technology doesn't live front and center in their lives. It's a complement, at times, but not their lifeblood.
[WARNING: GROSS GENERALIZATIONS AHEAD!!! Seriously, I love the Valley. Many of the world's most important technological (and human) innovations have emerged from the Valley. So take this post with a grain of salt. I just think we make too much of ourselves sometimes. It's important to remember that good companies make technology for customers. Not themselves. The latter kind go out of business..../WARNING]
Silicon Valley doesn't get this. Silicon Valley lives in a technology cocoon - a bubble that shields it from thinking about the real world, about real customers. Fortunately, the Valley has a few airports which provide the opportunity to take field trips to meet real people, but things like Flock make me think the Valley just doesn't get outside of itself enough.
As an example, drive down 101 and you're bombarded by technology billboards. Drive down any other freeway in America (or other countries that are unfortunate enough to have littered their freeways with advertisements, like we have) and you'll find advertisements that relate to real life: beer, divorce lawyers, and home builders. Sure, you'll find technology billboards here and there, but not the full-time fetish that plagues Silicon Valley.
And it is a plague. I believe the Valley is, contrary to popular belief, vastly inefficient because it wastes so much time and resources building technology that no one but a Silicon Valley developer could love (and even they have a hard time with much of the excess). This may well be why every major open source company today came from outside the Valley: MySQL (Sweden, though they've recently moved headquarters to the Valley, I'm not sure they're helped by it - perhaps not hurt, but not helped), Red Hat (North Carolina), JBoss (Georgia), SUSE (Germany). SugarCRM is on the upswing and may prove the exception to the rule...as will Alfresco.
Yes, we have Google, Yahoo!, and others. But these are a drop in the ocean compared to the man-hours and VC dollars that have been dumped into the Valley. And they still don't compare with IBM, Microsoft, and "yesterday's" software companies that have generated much more money and market value (though I believe they'll get there soon enough, and I'm expressly not saying that no good/great companies emerge from Silicon Valley. Obviously, they do).
Actually, Microsoft is a great example of a company that thrived because it grew up outside the Valley. Whatever one may think about Microsoft, it has done more than any other company on earth to lower the bar to technology adoption. Why? Because Microsoft makes technology (relatively) easy. For consumers, IT administrators, and developers. Microsoft recognizes that the average user isn't a Silicon Valley, highly paid, highly tagged/Flickr'd/blogged developer. The average user is...average.
Strange as it may sound, you learn this much more easily outside the Valley than within - it takes a lot of field trips to the world of the Luddites and Neanderthals to fully appreciate just how average the rest of us are. And how much more disposable income/IT budget we have to spend. You should try it some time.
[Let the flame war against Matt begin!]
Posted by Matt Asay on October 25, 2005 07:40 AM
October 24, 2005 | Comments: (0)
Technology innovation and differentiation
Following Matt's post on open source needing to be useful and not just cheap one key point came to mind regarding the Microsoft FUD that we are always commenting on. Many open source products have gone from being good enough to having true differentiation points. That's why the FUD targets areas that are not differentiation points like features and functionality, but rather metrics like cost of installation that are much harder to elucidate.
Replacement products generally take a bottom-up approach, moving from base requirements to par to better. Take for example online trading; It used to be that the high-end brokerage houses were the only choice because they had all the features. eTrade basically blew up that model and now has more features than the older brokers. Part of that is because they used technology to their advantage and part of it is because customers were ready for a change.
That was a long preamble to the WSJ 2005 Technology Innovation Awards.
Gold: 454 Life Sciences-Low-cost gene sequencing
Silver: Ecology Coatings-environmentally friendly coatings
Bronze: Alien Technology-RFID tags
Software: Agitar-bug finding tool
Posted by Dave Rosenberg on October 24, 2005 12:17 PM
October 24, 2005 | Comments: (0)
Arsenal, rising ticket prices, and open source
I'm reading Nick Hornby's Fever Pitch, Hornby's portrait of his own addiction to Arsenal Football Club (which was unfortunately mangled by Hollywood that turned it into a lame baseball movie. Who cares about baseball?). It's light reading, but a welcome intermission between Dickens and whatever comes next (probably more Trollope). And, of course, I share Hornby's love for Arsenal....
In the midst of the book, Hornby pulls back to talk about how the game is changing, and particularly the effect of rising ticket prices. As he notes in this quotation below, the big football clubs are effectively pricing away the very people that make the game so much fun to experience, which has its analogue in open source:
Arsenal and Manchester United and the rest are under the impression that people pay to watch Paul Merson and Ryan Giggs, and of course they do. But many of them - the people in the twenty pound seats [Note: This was written nearly 15 years ago. The seats are much more expensive now...though Giggs is still awesome.], and the guys in the executive boxes - also pay to watch people watching Paul Merson (or to listen to people shouting at him). Who would buy an executive box if the stadium were filled with executives?Not dissimilarly, while I applaud the merger of commercial and open source, I sometimes worry that too many executives will spoil the original flavor of open source. By this I don't mean the "community," because I think we've long misunderstood the 'community of communities' that open source truly is (thinking, instead, that it was one big blob of community). Rather, I mean the ethos of open source: giving back, collaboration, etc. We need to be careful.The club sold the boxes on the understanding that the atmosphere came free, and so the North Bank [at Arsenal's Highbury Stadium] generated as much income as any of the players ever did. Who'll make the noise now? Will the suburban middle-class kids and their mums and dads still come if they have to generate it themselves? Or will they feel that they have been conned? Because in effect the clubs have sold them tickets to a show in which the principal attraction has been moved to make room for them.
Posted by Matt Asay on October 24, 2005 12:02 PM
October 24, 2005 | Comments: (0)
AirTran should lighten up (Lesson for open source, as well)
The WSJ is reporting [Subscription required] that AirTran Airways may stop serving Anheuser-Busch products on its flights because of Bud Light ad that pokes fun at discount airlines. One representative quote from the commercial: "Sure we're concerned for our lives," the discount airline pilot/announcer says, "just not as concerned as saving nine bucks on a roundtrip to Fort Myers."
Pretty funny, actually. But not to AirTran and others (Southwest's spokesman said the ad was "Totally tasteless. To joke about safety is awful. It's out of bounds." I guess Southwest doesn't want to be compared to ValuJet, but isn't it fair game to make that association?
Regardless, unless these discount airlines see some truth in the ad, why do they find it threatening?
The same holds true for open source software companies. To the extent that such companies lead with the "We're cheap!" line, they deserve to be pilloried as 'less-than' competitors to proprietary software. That's why I like the substance (though not the tone) of Matthew Szulik's comment about selling value, not cheapness.
Though lots of companies are raising money on open source, they really should be raising money on the same thing that has always won over customers and generated profit: Value. Product innovation. Etc. If someone is promoting open source junk, it's still junk. We buy things because they're useful/valued, and not because they're cheap. Cheap is a bonus, as I wrote awhile back in IT Manager's Journal.
To the extent that open source companies focus on quality (as I believe Alfresco, MySQL, JBoss, SugarCRM, and others do), there is no need to fixate on price. That's a bonus that comes when the customer realizes that open source delivers equal or better (product) value, and at a lower price, too.
Microsoft gets this. That's why its open source moves (including the recent simplification of Shared Source licenses) are less about being popular and more about keeping the good the company already does, and making it better through a more efficient development model. But only when/where they can demonstrate that open source will benefit its customers.
It's what I like about Apple and Microsoft, both. As well as Google and every other successful software company: open source is treated as a useful means to an end (customer satisfaction), and not as an end in and of itself. That's mature thinking. These companies are customer "bigots," not license bigots. That's a tolerable and, indeed, admirable, form of "bigotry."
Posted by Matt Asay on October 24, 2005 11:44 AM
October 24, 2005 | Comments: (0)
IBM expands free access to patents
IBM announced that they would provide free patent portfolio access to the healthcare and education industries. The move is designed to get these industries building on open standards without fearing IBM's wrath. This follows the earlier 500 patent offering to open source.
IBM's patents could be used toward standardizing electronic health-record networks, the company said. In emergency rooms, for example, IBM patents could help create technology to speed the prioritization and authorization process in requesting patient information via Web services applications.Big Blue is also giving free access for standards-based applications in the education arena. IBM's patents, for example, could be applied toward technology that lets students living in remote areas access teachers, lesson plans and other resources.
More info in the SJ Merc
Posted by Dave Rosenberg on October 24, 2005 09:55 AM
October 24, 2005 | Comments: (0)
MySQL 5.0 is available for download.
Marten Mickos was kind enough to speak to my MBA marketing class this weekend and gave everyone a very honest view into how MySQL operates. He also noted that being CEO is sometimes a lonely job. Show him how much you love him by downloading MySQL 5.0
Posted by Dave Rosenberg on October 24, 2005 09:45 AM
October 24, 2005 | Comments: (0)
Merrill Lynch: CIOs taking a 'wait-and-see' approach to Vista
...and are taking an "install-and-see" approach to Linux. As reported by CNET:
According to a Merrill Lynch survey of 100 North American CIOs, two-thirds of businesses are waiting for more information before planning their upgrade schedule to the next version of the operating system, which is due out late next year.Just 8 percent said they are planning to upgrade in 2007. Twice that many are planning upgrades in 2008, while 5 percent said they plan to wait until 2009.
Meanwhile, just over half of those surveyed said they planned to increase their spending on Linux....
Posted by Matt Asay on October 24, 2005 09:42 AM
October 23, 2005 | Comments: (0)
BusinessWeek's Cold Realities For Novell is a very interesting piece on the state of Novell and the challenges the company continues to face.
Earlier this year, investors privately expressed concerns about Messman's ability to retain key execs and manage operations. Then on Aug. 25, the Waltham (Mass.) company surprised Wall Street with a 90% decline in quarterly profits, to $2 million, and a 5% sales slide, to $290 million. Critics quickly went public with their concerns and began calling for change at the top. "I think ultimately fresh leadership is needed," says analyst Jason Maynard of Credit Suisse First Boston (). Blum Capital, with a 5% stake in Novell, told the company in a letter that its strategy is sound, but "the question is whether the current management and board will execute."
The integration of Suse has been portrayed as a difficult task, and I think the question is if Chris Stone, who many thought was going to be CEO before a hasty resignation in 2004, could have done a better job? And for that matter, could anyone have led Novell upstream? Is Messman being blamed for a transition that was going to take this long anyway?
I never found Novell to be particular compelling, though I do appreciate their efforts to try be a Linux company. As an observer, it just never felt like Messman really got it, unlike Matt Szulik who seemed as much a believer as a business pragmatist. I think that replacing Messman would present further challenges for Novell but still probably needs to be done. I'm not sure the corporate culture could handle it, but I don't think the upcoming layoff of ~20% is going to be so great either. Especially in Utah where tech jobs are hard to come by.
If Novell can't regain its footing, it could represent a major setback for Linux. The software has gained considerable traction in corporations, with nearly a 25% share of the server operating system market, according to market researcher IDC. Yet customers and the computer makers who back Linux want two strong Linux distributors. And right now Red Hat Inc. () seems to be running away with the market. It had 63% of the Linux server market share in 2004, compared with just 20% for Novell. "We don't see Novell that much," says Timothy Yeaton, Red Hat's senior vice-president of worldwide marketing.
If I were Microsoft I would start trying to acquire Novell before all the talent leaves.
On a separate note, we had some discussion about the quality and one-sidedness of certain "journalists" covering open source. Cold Realities For Novell by Sarah Lacy is a great example of even-handed journalism, researching and finding the facts and letting the reader decide which side they land on-not prescribing one's own skewed opinion as fact and abusing the power associated with mainstream publications. Thanks Sarah.
Posted by Dave Rosenberg on October 23, 2005 04:40 PM
October 22, 2005 | Comments: (0)
O'Reilly's Open Sources 2.0 finally hits the bookshelves
I wrote a chapter for O'Reilly's Open Sources 2.0 last year, and it's finally hitting the shelves. Not sure as to the reason behind the delay, but I'm looking forward to getting a copy. No, not because of what I wrote (I've already read that), but because of the other interesting perspectives Danese, Chris, and Mark collected on the present and future of open source business. If this book is anything like the 1.0 rev (Open Sources: Voices from the Open Source Revolution), it will be well worth the price of admission. 1.0 centered around the disruptive development model; 2.0 focuses on open source as a disruptive business model.)
My chapter is called "Open Source and the Commodity Urge: Disruptive Models for a Disruptive Development Process." I've included the abstract and introduction below, to give a flavor for what the book offers.
Abstract: Open source hastens software's natural trend toward standardization/commodification. While technologically innovative companies will always find ample customer interest, the most important innovations for the next decade of software will come from business model innovation, mostly spawned by open source license requirements. Open source builds a new intellectual property regime centered on the source of code, not source code. Protection, in other words, shifts to "owning" the code creator, rather than the product she creates. Those business models that acknowledge this and leverage it will yield better profits than those that attempt a half-way embrace (or rejection) of open source.
And here's the Introduction to the chapter:
IntroductionWe are missing the point. Yes, open source imposes dramatic changes on the software industry and, yes, it is roiling the fortunes of many an established vendor. It will continue to do so, and at an increased pace. Yet despite the sometimes anguished, sometimes giddy reception that open source has provoked in the IT world, open source is not novel. It is not odd.
Open source is simply the software world's mechanism for becoming just like everyone else.
All the world's a commodity, or services to support and distribute commodities. This book that you are reading. The chair that supports you. The restaurant you will eat at tonight. Everything, including, increasingly, software, thanks to open source. Open source accelerates the natural progress of software toward commodification, or standardization.
It is critical that IT vendors understand this so that they can deploy (or fight, if they so choose) open source effectively, and more intelligently choose how and where to innovate. Open source does not destroy all value in software innovation; instead, it shifts the control point from the code itself to the creator of the code. In so doing, open source software will not pillage all closed source software. As in other industries, there will continue to be plenty of room for upmarket vendors (e.g., Whole Foods in grocery retailing; Starbucks in coffee; and Nordstom in retail clothing).
That said, there is no room for middling and muddling. Open source will commodify from the bottom-up while "upmarket" vendors will dominate "up-the-stack." Everything else will be a wasteland. Just as Safeway finds itself pummeled by Wal-Mart and Whole Foods so, too, will middle ground IT vendors find themselves grasping at a dwindling market opportunity.
Open source offers hope, but perhaps not for the reasons normally associated with it. Much has been made about the open source revolution, and with good reason. But perhaps the best reason has little to do with development of source code, and instead has much to do with distribution, marketing, and sales. In other words, what we thought was a software development methodology may have far more importance as a business strategy that undercuts competitors while driving down costs and shifting control to buyers. In such a world, those who understand and leverage open source commodification (or escape it) will thrive - everyone else will be marginalized into economic oblivion. Commodification, the highest stage of capitalism; open source, the highest stage of software.
Posted by Matt Asay on October 22, 2005 04:29 AM
October 21, 2005 | Comments: (0)
eBay: Driving revenues by driving them away
It's no secret that I'm baffled by the eBay acquisition of Skype (both by the demographic issues and by the rationale eBay used. It's not that the acquisition wouldn't make sense for someone (like a cable company, perhaps?), but rather that I just can't see my grandmother putting on a headset to chat with a Bulgarian merchant on Skype. But then, I've been wrong before. She talks to all sorts of people....
Meg Whitman, CEO of eBay, is on the record now with a dubious justification of the purchase. eBay will make gobs of money from Skype by...giving it away.
"In the end, the price that anyone can provide for voice transmission on the 'Net will trend toward zero."OK. I believe that. But what does it say about eBay's acquisition of a company that only makes money by charging users to make calls? (Even the open source advocate in me can't seem to get beyond the non-logic involved.)
Whitman's reply is that money will not be made on voice calls, directly, but rather subsidized through advertising and transaction fees. Maybe.
I'm with David Coursey on this one, however. I don't want free but ad-laden voice service. I'm actually kind of sick of having to wade through ads everywhere. Just charge me for it and leave me alone. For, as Coursey reminds us:
1. Nothing is ever free.I don't want ad-supported telephone service. I don't want someone else paying for my time spent on the phone, just as I don't like paying advertisers to spew ads to me on TV. (Hence, I don't watch TV. Unfortunately, I can't quite get out of buying their products....)
2. Somebody always pays, and whoever pays gets to choose.
3. You get what you pay for.
Let's call a spade a spade. eBay wanted to look cool again. Skype is cool. So, the thinking goes, eBay + Skype = a cool eBay (to match the "coolness" of Google, Yahoo!, etc.)
Wrong. Revenues are cool. Profits are cool. eBay pulled both in this quarter, neither because of Skype. Time to get back to your core business, eBay. It's not telephony.
Posted by Matt Asay on October 21, 2005 03:05 PM
October 21, 2005 | Comments: (0)
The future: Making open source software more usable
After my 9am meeting in San Mateo failed to materialize I ran into Tara from SugarCRM who mentioned that John Roberts was speaking at OSBC and she was wondering what I thought the next big open source talking points were. My feeling, and one that I have shared with Sugar in the past is that the whole "open source is moving up the stack" message is important, but I think we are past it. We need to get to a stage of where open source applications are the obvious choice. Sugar has gotten pretty far down that path, as has MySQL, and it's clearly happened with Apache and Linux, but with the rest of the enterprise application ecosystem we’re not there yet.
For the masses to adopt open source software it has be as usable and easy to install as the proprietary alternative. As an example, I downloaded OpenOffice 2.0 yesterday for Linux and received a whole set of RPM files that I had to remember how to install. Could I give that to a casual user? No. I can't even recommend people test-drive it in that way. It should be a brainless install, like how software is on the Mac. But OO.org is not a commercial company. Even when you download SugarCRM you need some other components, but the Spike/Sugar Stack makes installation much easier.
Making open source software easier to install and deploy should be a top priority if we expect enterprises to adopt it.
On a side note it seems like many open source startups get a little distracted by the whole "Open Source Community" aspect and spend too much time managing press and not enough time getting meaningful customers to go on record using their products. Take for example SourceFire, which did virtually no marketing and somehow ended up being acquired for $225 million. They focused on getting customers and listened to users and made their products more appealing. In contrast look at Matt's favorite-SpikeSource which made a big PR splash but have yet to easily explain what it is they do, and haven’t messaged who is paying for their services. I actually think SpikeSource does some cool stuff, but it's taken six months for me (and arguably a larger audience) to figure out what the hell it is.
Disclosure: Matt is on the SugarCRM advisory board and my wife uses the SugarOnDemand service.
Posted by Dave Rosenberg on October 21, 2005 12:17 PM
October 21, 2005 | Comments: (0)
Wikipedia: Maybe the masses aren't asses, after all
So, the news this week was that Wikipedia's founder admitted it has "serious quality problems." In part a response to Nick Carr's scathing rebuke of the rise of the amateur class, also picked up by InfoWorld.
Wikipedia being what it is, it actually has an encyclopedia entry called "Criticism of Wikipedia". There's something about that candor that you just have to love.
Actually, the more I think about it, the more I wonder how valid all of this criticism is. I'm no deconstructionist (spent all my grad school years fighting against such limp-minded buffoonery), but despite my initial going with the flow ("It stinks! I've smelled it!!") I'm wondering if the criticism is misplaced.
After all, just who, precisely, constitutes an expert in people, places, and things?
What makes a good encyclopedia? If I'm reading it today, I suppose I want a maximum of distilled, comprehensive facts on a given subject. [That said, I think I'd be equally interested in the subjective journal entries from those who were fought in the Battle of the Bulge as I would be about the date, location, etc. Actually, I'd be much more interested in the journal entry. Because that's where my humanity connects with theirs.] But if I'm reading it tomorrow (meaning, if I were an anthropologist stumbling across an encyclopedia from 100-1000 years ago), I think I'd prefer to read a collective bubbling up of a community.
In other words, I'd want to read what a community thought about itself.
This is, really, what "normal" encyclopedias do, but they make pretensions of being something more. They pretend, that is, to be authoritative. I guess I'm not really sure what that means anymore. Does it mean the cathedral or the bazaar? Which information really matters most? What the titled experts think, or what the collective experts think?
I think both - cathedral and bazaar - have their place in encyclopedias. But if I had to choose one, I think it would be the bazaar. Not because it's necessarily better information (though I do believe it will tend to be more comprehensive in scope, if the contribution community is wide enough), but because it's local information. This is a blessing, not a curse, of the rise of the commons.
I suspect that the real problem with Wikipedia is simply that it's still young enough that it lacks a suitably disparate and large community behind it. Assuming it provides even a modicum of value today (it does), then people will build on that modicum until it becomes ever more useful as it collects our communal knowledge. The masses may be asses (thus spake Alexander Hamilton), but it's better than the tyranny of the few over time.
Posted by Matt Asay on October 21, 2005 08:43 AM
October 21, 2005 | Comments: (0)
Jonathan Schwartz on Ajax: Sexy is as sexy does
Jonathan sticks a wrench in the Ajax cogs today, declaring (GASP!) that it is not manna from heaven, a sacred language divulged by the angels, and other such heresy. Shame on him for being rational. :-)
In a nutshell, Jonathan suggests that an application's language is immaterial to the value it conveys. The two are separable. Distinct. The value is in the...value.
As a software guy, here's a simple (though often irritating) rule behind user oriented software: The language in which a product is written has nothing to do with the value it conveys. Coming from the company that produced Java technology, that probably sounds a little odd. But it's a simple truth, especially when it comes to users: if the app's no good, it's no good, even if it's implemented in Java. Or PHP. Or Rails.Amen. We can be goofy about technology trends. But at the end of the day, Web 2.0/Ajax applications are only worth the end-user value they actually drive. And that has nothing to do with a pretty interface (though I'm a sucker for the superficial - I freely admit it), and everything to do with what users actually get from the application.For whatever reason, the first internet boom led a very broad developer audience somewhat astray with opportunity - no end of new software products were produced that were, in essence, simply old products either rewritten, or built to run through a browser. Granted, many succeeded. But at least as many (actually, way more) failed. Why?
Because rewriting an app simply to use a new toolkit isn't creating value for consumers. Creating an application or service that delivers unique value is what captures users. And the internet gave some developers a tremendous opportunity to deliver unique value - by radically simplifying basic networking, enabling connectivity and community on a truly global scale.
Posted by Matt Asay on October 21, 2005 07:22 AM
October 19, 2005 | Comments: (0)
Licenses, licenses everywhere, but not a bit of added value
So, the BIG NEWS today is that Microsoft came out with Yet Another License. Three more, to be exact. Supposedly (at least, according to Jason, whom I trust and respect), this is supposed to ensure that
all of Microsoft source code releases will be under consistent terms, and thus more easy to use and to understand. The licenses are each 1 page or shorter. They are written in simple terms that non-lawyers should be able to follow. They are also reflective of the most modern thinking regarding source code licenses within the legal community.Sure. OK. Who cares? (Well, Tim gives us some reasons to care, most of which I covered below. But then, I'm not Tim. Hearing it from him is important....)
To the best of my knowledge, open source has not been hampered in any way by its existing license base. We have too many (and now Microsoft says that it has shaved 10 Shared Source licenses down to three - that's progress, but toward what?), but what we don't really have is a communication problem. I'm all for making licenses simpler, but it's a fool's errand to believe that shorter = less legal wrangling. Lawyers are lawyers. They get paid by the hour.
So, I don't see these new licenses adding anything to the open source community. Why? Because they're licenses, not code. Code matters. Actually, let me take that back. Code matters very little. Community matters most.
So, a much bigger announcement would have been that Microsoft had decided to join the open source community in earnest. Stephe Walli has been hammering on this point for some time. Or, at least, Microsoft could have loaned Steve and JD to the community to interweave all the existing licenses.
Of course, if these licenses are meant to facilitate Microsoft's joining in with the community in a more meaningful way, then I'm all for them. I think it's silly to invent new licenses, but if it's what Microsoft requires in order to take baby steps towards deeper involvement, publish as many new licenses as you want. Because at the end of the day, Microsoft has got tens of thousands of developers, which developer capital is crimped and constrained by trying to channel all that energy into Office and Windows. Let them break free. Let them contribute.
Posted by Matt Asay on October 19, 2005 05:37 PM
October 19, 2005 | Comments: (0)
Microsoft misses the mark---again
At what point does Microsoft's PR team stop Alan Yates from speaking publicly? Yates, GM, information worker business strategy is hell-bent on making disparaging remarks about open source without facts or statistics to back them up.
"Clearly, it's changing the landscape of the technology industry for better and worse,"Some groups want to remove intellectual property completely, but Yates said commercial companies such as Linspire produce operating systems based on Linux, the most widely used example of open-source programs. He said such companies stand by their products and provide a balance between the open-source and proprietary sides, a middle ground of sorts
But Yates said government agencies and organizations need to understand the benefits of using open source and balance that against implementation costs. He said they need to understand licensing privileges, indemnification issues and an ecosystem of service providers, among other considerations.
I'm curious what Yates is referring to when stating that open source has changed the technology industry for the worse. Licensing and Indemnification? Weak. Implementation costs? Hardly. I'm not naive enough to think that open source is without fault. There are lots of issues. But the issues Yates brings out signify a lack of understanding of the technology marketplace and instead prove that Microsoft is still out of touch, and still resorting to sophomoric attempts at scaring people into avoiding open source.
One thing that I am seeing hands-on is that a box is a box. It doesn't matter if it's Linux, Windows, or Solaris, somewhere down the line you have cost associated with deploying and maintaining any system. Even though Windows is perceived as easier to setup the fact is that the dependencies and variables associated with a secure setup are actually more difficult than Linux is. FYI-Solaris has the best documentation.
I am currently going through a deployment of two new system sets, one LAMP based and the other .Net (that's right, I too am terrified.) All of the systems are being outsourced to a managed hosting provider, because we didn't feel comfortable managing the Windows servers without dedicated staff. Managed hosting for Windows (redundant 3-tier architecture) w/SQL server is three times the cost as it is for Linux with MySQL. Every hosting provider says the same thing-that Windows requires more support and management than Linux does. We currently have a series of Linux boxes in a collocation facility that have been up for almost 2 years. Even I think that's ridiculous.
Matt is going to take a shot at the new Microsoft licenses they announced today. It's not my area of expertise, but I will ask MS the same question I asked CA when they wrote a new license (or for that matter why they thought an open source Ingres was going to be meaningful); What's the point? There are too many licenses already, adding more to the mix just exemplifies the lack of understanding of the issues in open source--sorry, "shared source".
One more thing...
Since no one (except Jason) from Microsoft reads this site I hereby offer to Microsoft the services of Matt and myself to explain why they just don't get it. All we ask for are pre-release copies of Vista with WinFS and a large cash prize.
Posted by Dave Rosenberg on October 19, 2005 03:38 PM
October 19, 2005 | Comments: (0)
Bob Young resigns from Red Hat's board
Today Red Hat announced that Bob Young has resigned from Red Hat's board, apparently to focus on his publishing business (Lulu).
"In a funny way, my resignation is perhaps the finest compliment I can pay to everyone associated with Red Hat today, I have complete confidence in the future of the company," said Young.Surely that sentiment rings true - Bob's the sort of guy that would serve until he felt he was no longer needed.
Still, no Red Hat executive leaves without a strong nudge from Matthew (and there have been plenty of nudges over the past few years, including Kevin Thompson (CFO), Tim Buckley (COO), Hal Covert (CFO before Kevin), John Young (VP, Marketing), etc.). I wonder if Matthew was clearing space on the board, and I'd assume that Bob has actually been somewhat disengaged for awhile. I spoke with him on a panel in Toronto some time ago, and he was pretty focused on Lulu.
At any rate, he's an exceptional person, and I wish Bob well.
Posted by Matt Asay on October 19, 2005 06:04 AM
October 18, 2005 | Comments: (0)
If you ask Eben Moglen what he does for a living, he'll tell you he's a freedom fighter.
Columbia Law School Professor Eben Moglen wants to destroy the Federal Communications Commission. Not as some kind of terrorist act, but because technology is rapidly making it irrelevant.The agency might have made sense in the 1920s, Moglen says, when it was formed to assign specific frequencies to broadcasters so they wouldn't try to drown each other out by cranking up the transmitter power. But a new generation of intelligent radios, combined with equally clever computer networks, is making it possible for anybody to use the airwaves without interfering with anybody else.
Despite some factual mistakes it's an interesting read.
Posted by Dave Rosenberg on October 18, 2005 12:08 PM
October 18, 2005 | Comments: (0)
Matt Asay leaving Novell...revealing his biases (al fresco style)
I've taken a little flak for my posting on SharePoint. There I hype Alfresco. There's a good reason for that hype (I love the company and think it's going to be a major player in the ECM market), but there's an even more personal reason that starts November 1.
But let me backtrack a bit. Despite my false start the last time around, this time it's real. I'm leaving Novell.
This has nothing to do with Novell - I love the company, and it has been nothing but generous with me. I've spent a little over three years with the company, long enough to co-found our Linux Business Office with Chris Stone (then Vice Chairman), Al Nugent (then CTO, now running CA's Unicenter business unit), and Jeff Hawkins (my boss and VP of Product Management) (and recently disband it once its job was done, i.e., turning the company into a vibrant open source company), help build Novell's ties into the open source business community/ecosystem (MarketStart bein