Weekly Edition
Return to the Announcements page
| |||||||||||||
|
| |||||||||||||
Stanford professor on competing with free software
The Stanford Graduate School of Business is highlighting the work of one of its professors (Haim Mendelson), who is working on how proprietary software companies can be competitive with free software. "There are times, however, when the open source product gets to market first. In that case, the commercial vendor does well to enter the market with a compatible product and then invest in new product features to make its product compelling even though it costs morea strategy sometimes known as 'embrace and extend.' In this case, being open (or compatible) helps the commercial firm tap into the network created by the free product. Then, the commercial firm must compete by out-innovating the free product.?
(Log in to post comments)
Stanford professor on competing with free software Posted Sep 23, 2008 15:18 UTC (Tue) by jospoortvliet (subscriber, #33164) [Link]
Out-innovating the Free product, huh. Well, as the FOSS community matures,
I believe innovation will become its greatest strength. In other words - tough luck ;-)
Stanford professor on competing with free software Posted Sep 23, 2008 16:21 UTC (Tue) by proski (subscriber, #104) [Link] If a proprietary vendor finds that all ideas are quickly copied by free competitors, it might be more likely to use software patents to protect its innovations. Once proprietary vendors take position of underdogs, they could start employing nasty tricks they would otherwise eschew.
Stanford professor on competing with free software Posted Sep 23, 2008 16:41 UTC (Tue) by JoeBuck (subscriber, #2330) [Link] In application spaces where the target audience is software developers, free software has been extremely innovative. The proprietary software world never produced anything like Perl, just to give one example.But in apps where the end user is engaged in some specialized application that isn't software development, I've seen almost no innovation, and a lot of copying of proprietary apps, coming from the free software side. I don't think that this is inevitable, but it certainly seems common at this stage. I think it's because most free software projects start to scratch an itch, and if it's an itch that no free software developer has, it's not going to get scratched.
Stanford professor on competing with free software Posted Sep 23, 2008 18:12 UTC (Tue) by jordanb (subscriber, #45668) [Link]
You're right that the high profile applications copy the proprietary alternatives, but I think that that is more the result of pressure to be more like the propitiatory software. There has historically been a ton of innovation in Free window managers for instance. I'm here using Fluxbox with tabbed windows and paging worskpaces (that proprietary vendors got just last year).
Yet the popular, high profile GUI for Linux is the crappy Windows clone, GNOME, and whatever flunky window manager it's using now. Tools like Fluxbox don't get any sunshine because it's more important to be like Windows than to be good as far as the end user is concerned.
The reason why there is a lot of innovation in the areas of free software focused at nerds (you have to go way beyond 'development tools', look at how many innovative networking tools we've gotten) is because nerds reward innovation by using and promoting really innovative Free software, whereas what the public at large really wants is something that works exactly like their proprietary application, except for zero cost.
Stanford professor on competing with free software Posted Sep 23, 2008 22:28 UTC (Tue) by sbergman27 (subscriber, #10767) [Link]
"""
Yet the popular, high profile GUI for Linux is the crappy Windows clone, GNOME, and whatever flunky window manager it's using now. Tools like Fluxbox don't get any sunshine because it's more important to be like Windows than to be good as far as the end user is concerned. """
Most users have no need or desire for anything remotely resembling Fluxbox. Minimalism has its costs. A few people like the trade off. The vast majority of people don't. Fluxbox is in a niche because it is a niche product.
Stanford professor on competing with free software Posted Sep 23, 2008 22:50 UTC (Tue) by jordanb (subscriber, #45668) [Link]
I would hardly call Fluxbox 'minimalist,' it has quite a few more features than the GNOME WM, like tabbed windows.
And that's my point: you don't see those kinds of innovations in GNOME because the purpose of GNOME is to be like Windows. Fluxbox isn't a niche product because it's 'minimalist' but because it is innovative. Normal people don't want innovative, they want "just like Windows."
Stanford professor on competing with free software Posted Sep 24, 2008 10:41 UTC (Wed) by AlexHudson (guest, #41828) [Link]
If GNOME's purpose is to be "just like Windows" then it has pretty clearly failed, because it is nothing like Windows.
Thank goodness that it's not GNOME's purpose.
Stanford professor on competing with free software Posted Sep 24, 2008 17:33 UTC (Wed) by jospoortvliet (subscriber, #33164) [Link]
It's not, but innovation isn't either, so the parent has a point. I often say KDE 3.5.x and Gnome don't aspire more than a beautified and slightly improved windows 95, and I still think it's true. KDE 4 tries the first departures from that paradigm (and @jordanb: KDE 4 will most likely have tabbed windows, grouping etc as well - despite the fact you can hardly call it innovation, all these years behind Fluxbox, at least we try ;-))
Stanford professor on competing with free software Posted Sep 24, 2008 16:22 UTC (Wed) by ehabkost (guest, #46058) [Link]
It looks like you are looking only at the way the window manager behaves, ignoring all the rest. There is a lot of other stuff included in a desktop environment besides a window manager.
The way windows behave on Gnome really looks like Windows to me. Maybe because people don't really expect "just like Windows" on a window manager. But that doesn't mean users want "something like windows" on all other areas of Gnome (or any other desktop environment).
Stanford professor on competing with free software Posted Sep 23, 2008 19:47 UTC (Tue) by jospoortvliet (subscriber, #33164) [Link]
--> but it certainly seems common at this stage.
I think that's spot on. First you have to catch up, then when it's head to head, you can start to innovate. That takes some balls, and a clear and compeling vision. My pet peeve example here is the Free Desktop - the latest Gnome & KDE (3.5.x) are pretty much on par with Windows & Mac. Now it's time to start innovating, and at least the KDE project is clearly on track in that department.
Valid Line of Academic Research Posted Sep 23, 2008 16:46 UTC (Tue) by AnswerGuy (guest, #1256) [Link]
This strikes me as a perfectly valid line of academic research --- more under the category of economics rather than engineering.
However, of course, this particular snippet seems to focus on software as a
Most of the business that seems to have long term viability are cases were software is embedded into (or integrated with) a hardware offering (Cisco, Apple) ... or bundled with hardware (Microsoft), or intrinsically tied to proprietary (and frequently updated) data (Intuit), or focused in market niches where the customers are large enterprises (Oracle, and the big 3 or 4 EDA companies) or where the offering is primarily a service (Google) or, finally, where the support, and training, is the main offering (Red Hat, Novell, Canonical).
Naturally I'm over-simplifying in that statement. For example Cisco's primary market segment is also enterprise class customers, and at least one of Cadence's products is a hardware offering (Incisive "Palladium") I've heard that Intuit has recently focused on makign Quicken available as a web/AJAX service and phasing out the standalone versions, and so on.
The notion of "shrink wrapped" software ... as a product ... is mostly obsolete today and is certainly not a viable business model for the future.
Despite that observation it would be foolish of us to scoff at the idea that proprietary development groups, some of them, some of the time, will bring significant innovations to market faster than we will.
Valid Line of Academic Research Posted Sep 23, 2008 17:33 UTC (Tue) by ewan (subscriber, #5533) [Link] The notion of "shrink wrapped" software ... as a product ... is mostly obsolete today and is certainly not a viable business model for the future.It seems to work pretty well for the games industry, though I suppose you could consider that as similar to the Intuit ('proprietary data') case.
Valid Line of Academic Research Posted Sep 23, 2008 19:45 UTC (Tue) by jospoortvliet (subscriber, #33164) [Link]
yeah, games are more about the artwork and story than about thechnology...
Valid Line of Academic Research Posted Sep 23, 2008 23:13 UTC (Tue) by drag (subscriber, #31333) [Link]
I wish they were more about gameplay. But whatever.
I can see commercial open source games working out fairly well. Games generally use other people's code in them.. most of the very big games are built on the same small number of gaming engines from people like Valve, Unreal, Id, and similar. They generally don't use their own code.
Game content and artwork and whatnot isn't code and I would be happy to purchase a game that had a open source engine, but kept the game logic and artwork proprietary. Especially MMORPG were people will happily pay subscription fees to have somebody else maintain a persistent alternative reality for them. I don't know if it would be as profitable, but I expect that somebody like War of Warcraft could open source the engine they use and still make plenty of money.
Valid Line of Academic Research Posted Sep 24, 2008 11:55 UTC (Wed) by NAR (subscriber, #1313) [Link]
I think open sourcing the client of an online game is an invitation for the users to create "helper" applications. For an FPS it could be a sharpshooter-mod (misteriously all of the shots from the user are headshots), for an RPG it could be an automatic bot that trains or collects gold all night long, etc.
Valid Line of Academic Research Posted Sep 24, 2008 18:13 UTC (Wed) by drag (subscriber, #31333) [Link]
> I think open sourcing the client of an online game is an invitation for the users to create "helper" applications. For an FPS it could be a sharpshooter-mod (misteriously all of the shots from the user are headshots), for an RPG it could be an automatic bot that trains or collects gold all night long, etc.
Well that stuff happens on closed source games, too... With a pretty much constant basis. So I don't see open sourcing anything is going to make it any worse. It may make it easier, but it's not something I'd lose any sleep over.
I don't know if you know what somethingawful.com is, but they have a very interesting and successful way to control trolls on their forums. It's a joke website, so obviously your going to attract all sorts of odd trolls and whatnot. For a long time it was a constant battle using traditional methods like account banning, blocking IPs, and so on and so forth. In other words: a losing battle.
Well they struck on a wonderful method to control the trolls. One-time 10 dollar account fee and aggressive permanent banning with no appeals. They'll bankick you for very slight and often very silly reasons. You could then re-sign up as many times as you'd like, but it cost you another 10 bucks. If you behaved yourself then your 10 dollar commitment was all you'd ever need to be a member for life.
And it worked. A person may not agree with their particular brand of humor, but they do keep the trolls under control.
This works for two reasons:
2. Most trolls are going to be young male teenagers. Young male teenagers don't have credit cards and therefore can't get access to the system anyways. Not without help from their parents.
So instead of trying to enforce anti-cheat techniques through controlling the software, you instead manage identities. Say a PKI-style way to manage accounts. A person's 'game password' isn't something you store on your server, it's the decryption key to their private key. Then that private key is tied to their identity for the 'universe' you have setup for the game. Could be everything from keeping track of account statuses, to name ident stuff in chatrooms, to forums or whatever else you have tied up in the game.
Then in that way a person's good name becomes valuable beyond just how the people in the server your playing perceive you. If your part of a clan or if your involved in competition or want to keep track of your ranks or high scores then you'll have to protect that identity. If you lose control of it or you piss off too many server administrators (say if 5 independent server admins perma-kick you then you lose the account) then you can loose it and have to start your game over from scratch, plus pay the modest fee for another account creation.
This doesn't even have to be for a for-profit game either. The fee would just pay for the costs of managing account servers. Then other, independent, people putting up gaming servers could choose whether or not to tie their servers into the ranking systems.
rule number one: if they can't afford it, get them to pirate it Posted Sep 23, 2008 16:34 UTC (Tue) by dmarti (subscriber, #11625) [Link] "One segment is used to 'seed' the network so the product can achieve critical mass, which makes it more attractive to buyers in other segments who then end up choosing the more popular product over the free product."Often the seed segment is the "pirates" -- it's like an academic, emerging market, and microenterprise discount program but without the transaction or fulfilment costs.
Same answer to How Do You Compete Against Microsoft Posted Sep 23, 2008 23:39 UTC (Tue) by dkite (guest, #4577) [Link]
Provide something they don't.
Or how would one compete against Walmart? Sell what they don't.
There are very large holes in the free software stack. Free software has been
Another way to compete is to co-opt. Look at Apple's offerings. They took a
What is nice about even the question is that anyone who wants to stick around
Derek
Same answer to How Do You Compete Against Microsoft Posted Sep 24, 2008 0:38 UTC (Wed) by rahulsundaram (subscriber, #21946) [Link]
Atleast for khtml, I think KDE pretty much dropped the ball. I wouldn't blame Apple for WebKit and certainly don't see any "abuse". They have done a decent job with it.
khtml and Apple Posted Sep 24, 2008 8:28 UTC (Wed) by morhippo (subscriber, #334) [Link]
I think you are mistaken. Apple took khtml and forked it. Then they dumped huge patches on the khtml developers, which they had previously developed in a closed development environment without any input from khtml. Webkit as such was and is no longer compatible as khtml replacement. Initially, Apple made no effort to cooperate with the khtml developers.
khtml is quite usable and many improvements in webkit have been integrated.
I only stopped using konqeror because KDE4 network access does no longer work behind Fritzboxes anymore (some buggy IPv6 requests cause DNS to fail) and some KDE developers take the point of view that a new router is the best fix (even though the allegedly buggy routers work with all linux programs I have except KDE4). This bug is the most glaring bug in KDE4 and the most annoying one.
khtml and Apple Posted Sep 24, 2008 15:05 UTC (Wed) by michaeljt (subscriber, #39183) [Link]
If that is true, I fear that KDE use will drop dramatically in Germany.
khtml and Apple Posted Sep 24, 2008 19:44 UTC (Wed) by johnkarp (guest, #39285) [Link]
According to your own link:
* The problem can be reproduced easily at the command line, with no KDE/Qt involvement whatsoever;
khtml and Apple Posted Sep 25, 2008 9:12 UTC (Thu) by morhippo (subscriber, #334) [Link]
I think it is a shame that KDE had two major stable releases (4.0 and 4.1) which were unusable with very customary router software. The problem was known since the early betas, but has not been prioritized. I may have nostalgic memories, but I cannot believe this would not have happened in earlier release cycles.
A router update is not an option for many of us, as there are no updates for older routers available. The "buggy" hardware works with every other software on this planet, only KDE is affected. Do you really think that thousands of people should change their hardware, just because KDE sends out simultaneous IPV6 and IPV4 dns requests, even though my system is IPv4 only and all configuration options tell the system that this is so?
I fear that glaring usability bugs (i.e. KDE is unusable because of it) like this lead to a bad reputation of KDE, even though it is better than Gnome regarding its technical foundations.
|
|
||||||||||||