Kuro5hin.org: technology and culture, from the trenches
create account | help/FAQ | contact | links | search | IRC | site news
[ Everything | Diaries | Technology | Science | Culture | Politics | Media | News | Internet | Op-Ed | Fiction | Meta | MLP ]
We need your support: buy an ad | premium membership

[P]
Bruce Perens To HP: Open Source Isn't The Answer

By Carnage4Life in News
Wed Mar 07, 2001 at 10:19:28 AM EST
Tags: Software (all tags)
Software

In a ZDNet article on Hewlett Packard's recent decision to retire OpenMail Bruce Perens is quoted as saying 'I did not feel that we could support the OpenMail development team with an Open Source product -- licensing income would probably diminish to an unprofitable level' in reference to his opinion on whether HP should Open Source OpenMail. Bruce Perens clearly wanted to avoid what has become the situation with AOL and Netscape where they expected a majority of development to come from external sources but ended up footing the bill for most of the development while still having to give away the source.


Bruce Perens also aptly points out the fact that simply Open Sourcing a product does not guarantee that anybody will work on it as has been evidenced by the poor reception of HP's Open Sourcing of E-speak. It should be noted that several notable Open Source evangelists including Larry Augustin, CEO of VA Linux and David L. Sifry, Linuxcare's CTO are pressuring Bruce Perens to convince HP to Open Source OpenMail.

Sponsors

Voxel dot net
o Managed Hosting
o VoxCAST Content Delivery
o Raw Infrastructure

Login

Related Links
o ZDNet
o ZDNet article
o retire OpenMail
o E-speak
o Also by Carnage4Life


Display: Sort:
Bruce Perens To HP: Open Source Isn't The Answer | 22 comments (18 topical, 4 editorial, 0 hidden)
Here's why it's not open source. (4.50 / 8) (#2)
by rebelcool on Tue Mar 06, 2001 at 11:37:45 PM EST

"So why doesn't HP simply release the code to the public and allow it to sink or swim? "When a company decides to release existing proprietary code as Open Source, the show-stopper is almost always the other parties outside of that company who are involved," Perens said.

"Such parties become involved through patents that have been licensed, proprietary code that has been produced by a third party and embedded into the product, and existing contracts relating to the product that have been entered into with customers or other vendors. We don't know how big this sanitization project is yet, if it's bad, it could cost millions," Perens said."

In other words, the software uses tools and code from other companies. They would have to get permission from all those other companies to open source the project, which is a most unlikely thing to happen. Thus, before HP could open source it, they would have to basically write these tools over from scratch on their own. This would cost millions to do, and speaking from a business standpoint, it's simply not worth it to spend millions just to release the product for free. If it were my company, I wouldn't do it either.

COG. Build your own community. Free, easy, powerful. Demo site

Why not (2.66 / 3) (#3)
by enterfornone on Tue Mar 06, 2001 at 11:41:56 PM EST

Just strip the third party code, dump what's left on Sourceforge and let people pick it to pieces. Releasing a bunch of code that doesn't compile is better than just letting it die.

--
efn 26/m/syd
Will sponsor new accounts for porn.
[ Parent ]
heh, is it? (5.00 / 1) (#4)
by rebelcool on Tue Mar 06, 2001 at 11:47:14 PM EST

Even stripping the third party code would be extremely difficult. If the third party stuff is deeply ingrained within it, you would have to go line by line and find and remove all references. Honestly, you'd be left with a gutted worthless mass of code. If plans are to discontinue the project, you'd still have to pay the labor costs of removing the third party stuff. Programmers are expensive :) It's not worth the several thousand it would cost to do such a thing.

And before you say "but its not that much compared to how much they make in a year!" take a basic economics course. 10 thousand dollars is 10 thousand dollars they don't *have* to spend. Companies are profit maximizing beasts, and this would be something that costs alot for zero return. (Actually, judging from the size of such a piece of software, I can imagine stripping it down would cost well over 100K)

COG. Build your own community. Free, easy, powerful. Demo site
[ Parent ]

good point (4.00 / 1) (#6)
by enterfornone on Wed Mar 07, 2001 at 12:02:52 AM EST

I would have thought that they would keep thrid party code in seperate libraries or at least seperate directories in the source. But if they've just mixed it in then I guess it would be a hassle.

--
efn 26/m/syd
Will sponsor new accounts for porn.
[ Parent ]
it probably is... (4.00 / 1) (#7)
by rebelcool on Wed Mar 07, 2001 at 12:10:28 AM EST

however, you would still have calls to the various classes, and perhaps other dependencies that *are* in-line. And of course, there could be other contractual obligations which we are not privy to.

COG. Build your own community. Free, easy, powerful. Demo site
[ Parent ]

Leave the calls to the non-free components... (none / 0) (#21)
by Per Abrahamsen on Sat Mar 10, 2001 at 08:58:48 AM EST

...if they can release the rest of the code in an unusable state, it will be no worse than not having any code. And there just _might_ be someone (and old employee or user) willing to try to make it useful.

Of course, if the third-party code isn't clearly separated, or there are strange contractual obligations, it can be too expensive.




[ Parent ]
The problem is that there are likely NDAs (none / 0) (#11)
by Anonymous 242 on Wed Mar 07, 2001 at 08:22:56 AM EST

The project I work on uses one tool kit that has a very strict NDA attached. Exposing the API by means of publishing source code that has calls to the tool kit would violate the NDA. We're not talking about off-the-shelf software here, we're talking about third-party products that cost tens if not hundreds of thousands of dollars to license. With that kind of price tag, these vendors can afford some very expensive lawyers and if HP opens up OpenMail, you can bet these companies will be going through the code with a fine-toothed comb to see if anything that violates their agreements with HP was left in.

[ Parent ]
And this is surprising because? (3.81 / 11) (#9)
by GiTm on Wed Mar 07, 2001 at 06:32:38 AM EST

Open source is a good thing in my opinion, but I can certainly see why it's not a good thing for a lot of businesses.

Because OpenMail is already a HP product in most peoples minds these users are going to look to HP for support. Because it originated in HP people doing additional development are going to look to HP (specifically the original authors) for help. If HP could dump the source on SourceForge (minus any licensed/patented code) and pretty much forget about it I'm sure they would - net cost = 0, net gain = kudos for making the product available.

Unfortunately this won't happen - if HP just make the source publicly available and refused to have anything more to do with it then it would cost them in terms of negative publicity. If they just discontinue the product they are like any other company that discontinues a product and nothing special.

Open Source zealots tend to think that as soon as something is OS'd that hundreds of people are going to jump on it and turn it into a killer application or something. I can tell you from personal experience that this is NOT the case. There are far more consumers out there than there will ever be producers - and, even if the product is free, consumers assume that the producers accept some sort of responsibility for the product.

If Linux had started with Linus posting his original message to comp.os.minix - "Hey - it really sucks that we have to buy 'Operating Systems - Design and Implementation' to be able to get Minix, let's design and write a POSIX system from scratch" we would probably still be arguing about what should be in it. The fact that he did enough work to start with a (mostly) usable system certainly affected the number of people that were willing to contribute to it's success.

In my mind this is the biggest flaw in the famous 'Temple and the Bazaar' document by ESR. He based the document on the experience he had with an existing body of code that already had a reasonable amount of users. Have any of you tried to start up an open source project from scratch expecting that at least a few people would join you in developing the project? From personal experience I can tell you that most of the time it doesn't happen.

Have a browse of the projects on SourceForge - most of the projects started by an individual remain dormant and never progress - a single lone developer in the list. Projects that are started by a group of friends tend to remain active at least until they reach late beta and then seem to attract far more interest. Projects like Quake or Doom (I'm picking on games because that's my area of interest) - which are complete working products to start with will attract a large number of contributers from the word go.

Part of the reason is because people tend to over estimate their abilities. Another major part of the reason is because people believe that they will get some sort of assistance by like minded people.

I started a project over a year ago, I got a few expressions of interest but all of those people wanted to wait until I was ready (ie: had done most of the design work and initial coding) before they were willing to contribute.

Yes, this is a rant - but as a result of these experiences I don't even bother to check code into the SourceForge CVS server anymore. I'll keep it on my internal server until I'm happy with it (and I have something that at least runs enough for people not to like certain aspects and offer to help improve it) before I could be bothered to share it with anyone else.

Am I jaded by all of this and therefore somehow biased? Yes. Do I feel ripped off by the hype and expectations raised by Open Source advocates? Yes. Open Source is not a silver bullet - it seems to me that the Open Source methodology is an excellent way to maintain and enhance existing works but falls far short in the area of promoting new, innovative and risky software projects. It depends too much on the initial architecture, design and protoype development done behind closed doors.

Look at some major open source projects (almost the poster children) - GNOME & KDE, busy implementing features available in a commercial product for years (yes, that most hated of operating systems by the evil empire) - and doing them in a non-compatible, non-co-operative manner, even though both products work on the same platform. StarOffice, AbiWord and KOffice - once again busy trying to at least keep up with the features available in a commercial product. Mozilla (ex Netscrape) - a product abandoned by it's parent company in a sad attempt to extract sympathy from a community that already didn't like its competitor (Nothing against the people that have put in a lot of hard work for something they believed in but really Konquerer has come a lot further in a lot shorter time without anywhere near the hype and Opera has always been up there in terms of official standards compliance - even though it's not OS).

And the king, the saint, the media representative of Open Source - Linux. A re-implementation of 20 to 30 year old technology. If you really want a good example of a progressive and innovative open source operating system have a look at something like Atheos.

End of rant - asbestos suit in place.
--- I have nothing funny to say here.
Did you read The Cathedral And The Bazaar? (4.25 / 4) (#14)
by Carnage4Life on Wed Mar 07, 2001 at 10:32:12 AM EST

Am I jaded by all of this and therefore somehow biased? Yes. Do I feel ripped off by the hype and expectations raised by Open Source advocates? Yes. Open Source is not a silver bullet - it seems to me that the Open Source methodology is an excellent way to maintain and enhance existing works but falls far short in the area of promoting new, innovative and risky software projects. It depends too much on the initial architecture, design and protoype development done behind closed doors.

It looks like you never actually read the Cathedral and the Bazaar. In fact there is a chapter called Necessary Preconditions for the Bazaar Style that specifically states
It's fairly clear that one cannot code from the ground up in bazaar style [IN]. One can test, debug and improve in bazaar style, but it would be very hard to originate a project in bazaar mode. Linus didn't try it. I didn't either. Your nascent developer community needs to have something runnable and testable to play with.

When you start community-building, what you need to be able to present is a plausible promise. Your program doesn't have to work particularly well. It can be crude, buggy, incomplete, and poorly documented. What it must not fail to do is (a) run, and (b) convince potential co-developers that it can be evolved into something really neat in the foreseeable future.
ESR never stated that Open Source would be a way to create original, innovative projects nor did he claim that it was a good way to start a project from scratch. Even though your criticisms of ESR's paper seem to be unfounded, the paper is flawed and thus make some assumptions. Here's a good critique of it called Another Look at the Cathedral and the Bazaar.


[ Parent ]
not sure of that (4.00 / 2) (#15)
by streetlawyer on Wed Mar 07, 2001 at 12:57:43 PM EST

ESR never stated that Open Source would be a way to create original, innovative projects nor did he claim that it was a good way to start a project from scratch.

Well, yes and no. He seems to have forgotten what he said on a number of occasions particularly here, where he seems to be qualifying that statement to the claim that individuals have the ideas, but that under the Bazaar model, they can immediately attract a large and enthusiastic group of developers, precisely in the way in which it didn't happen to the original poster. Key quote:

If what it takes is one person with one good idea, then a social milieu in which one person can rapidly attract the cooperation of hundreds or thousands of others with that good idea is going inevitably to out-innovate any in which the person has to do a sales job to a hierarchy before he can work on it without risk of getting fired.
A bit ludicrous; Raymond appears to be saying at this point that it is more difficult to do a sales job to one person in a hierarchy, than to do a sales job to a mass market of thousands, which is not supported by many marketeers (of course, Raymond would never deign to talk to a marketeer about marketing, just as he is happy to put "anthropologist" in his profile without being one). Raymond also appears to be ignoring the problem quite horrifically in this bit:
That, however, is a negative point. The reader would be better served by a positive one. I suggest, as an experiment, the following;
  1. Pick any cloded-source operating system competing with Linux, and a best source for accounts of current development work on it.
  2. Watch that source and Freshmeat for one month. Every day, count the number of release announcements on Freshmeat that you consider `original' work. Apply the same definition of `original' to announcements for that other OS and count them. If your definition is ``I know it when I see it'', that's not a problem for purposes of this test. [streetlawyer editorialises: of course it is! but Raymond would never talk to a sociologist about observer bias, just as ,etc, etc]
  3. Thirty days later, total up both figures.
The day I wrote this, Freshmeat carried twenty-two release announcements, of which three appear they might push state of the art in some respect, This was a slow day for Freshmeat, but I will be astonished if any reader reports as many as three likely innovations a month ion any closed-source channel.
Anyone care to guess what happened to the twenty-two projects Raymond spotted? If you compare fully-fledged products rather than half-baked v0.0.01a pre-alphas, the figures look about the same.

--
Just because things have been nonergodic so far, doesn't mean that they'll be nonergodic forever
[ Parent ]
Lots of poor reasoning in that article (none / 0) (#16)
by Carnage4Life on Wed Mar 07, 2001 at 09:38:58 PM EST

A bit ludicrous; Raymond appears to be saying at this point that it is more difficult to do a sales job to one person in a hierarchy, than to do a sales job to a mass market of thousands, which is not supported by many marketeers (of course, Raymond would never deign to talk to a marketeer about marketing, just as he is happy to put "anthropologist" in his profile without being one).

I didn't interpret it quite that way. It seems more like he is saying that it's easier to convince a bunch of random hackers on the internet that your idea is innovative and is worth working on than it is to convince a corporate hierarchy. I actually agree with him.

It is a lot more likely that I can find some random hacker on comp.lang.c++ or Slashdot who agrees that a garbage collected C++ is a good idea but convincing my boss that we should devote resources to that would be nigh impossible.

On the other hand the reward for being able to convince your boss to devote resources to your innovative idea is usually worth more than convincing ten random people on the internet that your idea is cool.

The rest of my criticisms deal with Raymond's comments in his article
The GNOME project (to pick one of many) is pushing the state of the art in GUIs and object technology hard enough to have attracted considerable notice in the computer trade press well outside the Linux community.
Can this be the same GNOME project which is headed by Miguel De Icaza who has proclaimed publically to have gotten most of GNOME's design cues from Microsoft?
The day I wrote this, Freshmeat carried twenty-two release announcements, of which three appear they might push state of the art in some respect, This was a slow day for Freshmeat, but I will be astonished if any reader reports as many as three likely innovations a month on any closed-source channel.
I'm sure I see more than three hype-filled press releases for completed software products that push the state of the art in some way per month. Using Freshmeat as a judge of innovation in Open Source and trying to compare it to a closed-source channel (what is that anyway?) is ridiculous.


[ Parent ]
RE: Did you read The Cathedral And The Bazaar? (4.00 / 1) (#17)
by GiTm on Wed Mar 07, 2001 at 09:41:45 PM EST

I have read the Cathedral and the Bazaar, I probably should have re-read it before posting the comment though. I tend not to do a lot of detailed research for a comment - after all it's not a full article.
ESR never stated that Open Source would be a way to create original, innovative projects
Actually he does. In the [IN] link in the 'Necessary Preconditions for the Bazaar Style' leads to a paragraph containing the following text.
there is so much innovative work going on in open source that one is spoiled for choice. The GNOME project (to pick one of many) is pushing the state of the art in GUIs and object technology
To me that seems to imply that the Bazaar/Open Source model naturally leads to innovation and invention.

In subsequent discussions of the document ESR always seems to qualify his statements in such a way as to imply that the Bazaar model is a good method to start projects. Some examples are given in another post in this thread by streetlawyer.

Overall the OS community tends to propogate this belief - and as a result many potential valuable future contributers to the community become jaded. After being told how wonderful the model is and failing to attract any interest it appears to these nascent developers that they are not part of the clique. In the long term this cannot be a good thing.

My original comment contained a critique of the ESR document but that was far from the whole thrust of the comment.
--- I have nothing funny to say here.
[ Parent ]
Missing some fundamental concepts (none / 0) (#18)
by PresJPolk on Wed Mar 07, 2001 at 10:32:30 PM EST

Open Source zealots tend to think that as soon as something is OS'd that hundreds of people are going to jump on it and turn it into a killer application or something. I can tell you from personal experience that this is NOT the case. There are far more consumers out there than there will ever be producers - and, even if the product is free, consumers assume that the producers accept some sort of responsibility for the product.

So because HP won't get free development, then they should strand all of their existing users? You do know that Open source software and Bazaar development are two distinct concepts, right?

In my mind this is the biggest flaw in the famous 'Temple and the Bazaar' document by ESR. He based the document on the experience he had with an existing body of code that already had a reasonable amount of users. Have any of you tried to start up an open source project from scratch expecting that at least a few people would join you in developing the project? From personal experience I can tell you that most of the time it doesn't happen.

OK, obviously you don't. Anyone who starts an "open source" project with the idea that other will join does not understand the difference between the open source concept and bazaar development.

I, personally, work on several projects. One of them, Noatun, is very much a bazaar-style project. Charles Samuels wrote the core, and when I saw it and I liked it, I began writing plugins, and later hacking on the core, too. Later others joined. That's exactly the model Eric Raymond presented.

Another project I work on it Kit. There are only two people who have worked on Kit: me, and someone else who as just joined up. It's open source, giving full freedom to its users, but I'm not getting, nor am I seeking, a wide range of help in the form of code. I'll look at a patch if I receive it, but I won't feel bad if I don't.

Both Noatun and Kit are open source. Both are not developed Bazaar-style. So don't confuse the two concepts.

Am I jaded by all of this and therefore somehow biased? Yes. Do I feel ripped off by the hype and expectations raised by Open Source advocates? Yes. Open Source is not a silver bullet - it seems to me that the Open Source methodology is an excellent way to maintain and enhance existing works but falls far short in the area of promoting new, innovative and risky software projects. It depends too much on the initial architecture, design and protoype development done behind closed doors.

OK, now that our last concept is settled, maybe you should rethink the rest of this ... rhetoric. Open Source software is a freedom customers demand, it's not a development model.



[ Parent ]
Re: Missing some fundamental concepts (none / 0) (#19)
by GiTm on Thu Mar 08, 2001 at 01:22:23 AM EST

Yes, I was missing some fundamental concepts. Re-reading the Cathederal and the Bazaar (sorry about the incorrect naming in the original post) and the references in the follow up posts the distinction between 'Open Source' and the Bazaar model is a lot clearer now.

Unfortunately, in most references and discussions on this site and other media the two are very closely tied together - and quite often made out to be the same thing. This leads to the confusion for the majority of people (of which I was one) - which would lead to the problems for HP that I described in the original post. It seems to be more about perception than anything else.
--- I have nothing funny to say here.
[ Parent ]
Maybe there are times... (3.00 / 3) (#10)
by Rasvar on Wed Mar 07, 2001 at 08:07:18 AM EST

when a piece of software has just outlived is usefulness and needs to be retired. So HP does not want to just release the source code to openmail. That is their perogative. It is not as if it is such a mindnumbingly great product. It has gone through its life cycle. I used and supported it for years. It has since been supplanted by other products that are just easier, more efficient and better. Why does everything have to be opensourced?

A case in point (3.40 / 10) (#13)
by hardburn on Wed Mar 07, 2001 at 09:18:17 AM EST

For all those who beleive that Free Software and Open Source are mere synonyms, this subtely shows how they differ. Open Source says that, in many cases, there is a practical value in not closing the source. If there is no such value, don't bother. Thus, it is very pragmatic.

On the other hand, Free Software says that it is moraly wrong to not reveal the source (and allow people to modify it, redistribute their changes, etc.). A Free Software advocate would see this and say that it should be released with source whether there is value in it or not. Maybe somebody will care enough to look at it, maybe not. The important thing is that it is there.


----
while($story = K5::Story->new()) { $story->vote(-1) if($story->section() == $POLITICS); }


Concerning E-speak (2.00 / 1) (#20)
by Justin Goldberg on Thu Mar 08, 2001 at 08:20:32 AM EST

Why would it be popular when you have to sign up with your e-mail address just to download the source code? Not very conducive to openness, IMHO

Something fishy is going on here. (none / 0) (#22)
by erotus on Mon Mar 12, 2001 at 05:05:15 AM EST

First and foremost read this article from linuxworld. I'll give you a few highlights from this article and from another article from infoworld where HP earns the "Bury the Gold" award. Here is what Nicholas Petreley had to say about HP from the second link:

"The "bury the gold" award goes to Hewlett-Packard, which makes HP OpenMail 6.0. HP is sitting on a potential gold mine, but it is afraid to start digging. OpenMail runs on a variety of platforms, including Red Hat Linux, Sun Solaris, IBM AIX, and HP-UX. OpenMail is not only a drop-in replacement for Microsoft Exchange, it is faster, more scalable, more stable, less expensive, and performs some tasks more intelligently than Exchange. But HP won't market OpenMail as an Exchange replacement. HP is petrified of what Microsoft might do in retaliation."

Interesting isn't it? HP has a product that is a drop-in replacement for MS Exchange, runs on various platforms, and would hurt the MS monopoly. What is stopping HP? Is it really fear of retaliation? Here is a blurp from the linuxworld article:

"HP created an Exchange-killer and then hid it under a rock in order to protect its relationship with Microsoft. Now, ask yourself these simple questions. If I'm right in saying that HP was afraid to sell the OpenMail division, then where would HP find the courage to open source the product? Which is more dangerous to Microsoft: a pitifully marketed commercial product called HP OpenMail, a new company with a well-marketed commercial product formerly called OpenMail, or a free and open source version of OpenMail?"

Hmmm, Petreley makes an intersting point here. Either HP is scared or they're in bed with MS. I have a friend who works at a company that is involved with manufacturing. At their plant they use highly sophisticated robots and monitoring equipment. They use products from Agilent which is a spinoff company from HP. These products help them do monitoring and the Agilent products used to connect to a unix backend for reporting. Now, the new Agilent products only connect to an NT backend. What happened? Seems to be the monopoly forces at work again.

Finally, it seems that HP tries to kill it's best products despite the fact that these products are profitable. Openmail is the best current example. The other, is a system called the HP 3000. Is HP stupid or do they enjoy killing off good products? Is MS influencing them? Obviously so. Read the link I gave you and see what you think. Many people I've spoken to about Openmail, Agilent, and other HP deals leads me to believe that something dirty is going on behind the scenes.

Bruce Perens To HP: Open Source Isn't The Answer | 22 comments (18 topical, 4 editorial, 0 hidden)
Display: Sort:

kuro5hin.org

[XML]
All trademarks and copyrights on this page are owned by their respective companies. The Rest 2000 - Present Kuro5hin.org Inc.
See our legalese page for copyright policies. Please also read our Privacy Policy.
Kuro5hin.org is powered by Free Software, including Apache, Perl, and Linux, The Scoop Engine that runs this site is freely available, under the terms of the GPL.
Need some help? Email help@kuro5hin.org.
My heart's the long stairs.

Powered by Scoop create account | help/FAQ | mission | links | search | IRC | YOU choose the stories!