Sun wikis

Sun is deploying wikis as a public extranet on wikis.sun.com. They're using Confluence from Atlassian, a product I've used in two real life implementations of an entreprise wiki ([shameless plug] and with which I have significant experience now [/shameless plug]). Tim Bray has a post about the launch and wonders on his "wiki ranch" if he can make it more open to external contributors.

Meanwhile, I reacted on Martin's comment on Tim's blog:

Too bad Sun does not use free tools, like MediaWiki, for the internal Wiki. I manage a high-school reunion wiki, and although the syntax is hard to grok for non-techies, it is flexible and sooo open! It allows me to automate the registration process using some simple Python code between a Google Spreadsheet and the MediaWiki site, hosted freely at http://www.wiki-site.com. Libraries used: Google API (http://code.google.com/) and http://meta.wikimedia.org/wiki/Using_the_python_wikipediabot

If I had choosen other closed Wiki offering I could not have done all those nice automatic features!

Martin's comment is so typical. The real cost of a wiki isn't in the license, but in its development and growth. I've got several real life exemples where it would have cost several times MORE to use MediaWiki than to pay a license for Confluence and use it out of the box (that's what Sun uses too, and my little finger tells me it's not just a coincidence!). Most open source software (and MediaWiki in particular) absolutely suck at end-user features such as a really working WYSIWYG editor (no, wiki syntax isn't any "simpler" for casual users than HTML, it's just Fucking Ugly Code™ for them), or usability. And let's compare the complexity and cost of maintaining a bunch of "free" spaghetti code (bricolage!) assembled by a geek hobbyist to a product that comes with enterprise support and can be managed by non developers, etc., etc.

Sorry, but for entreprise wikis, the available most open source options are simply not up to par with the commercial offerings. I don't take a cut in selling licenses, and my clients perfectly understand that it's in their interest to pay a license if it's less costly for them than customizing a "free" software that they won't be able to maintain without endlessly paying a developer.

And a last note on MediaWiki, which fames derives only from Wikipedia, it's made with only one goal in mind: powering Wikipedia, period. Just take a look at the features (and lack thereof) and, most importantly, its documentation. One can only wonder, in front of such an horror: either the developers suck at documenting their code (too bad for people who make a tool for an encyclopedia) or the software isn't suitable for such a task. In either case, it's bad.

9 Comments

Agree. Good to see some push wikis forwards for the sake of adding value.

Furthermore, one may track simple numbers: Is the adoption of the wiki by the employees viral?... or not? Enterprises should congratulate themselves: This is just measurable.

And definitively, sometimes Confluence does well in respect to these numbers :-)

Hi Padawan,

You describe Martin's comment as typical, but I think that your comment is also very typical.

"than customizing a "free" software that they won't be able to maintain without endlessly paying a developer"

I'm perfectly fine that you are questioning the relative quality of Open Source Wikis vs Closed source Wikis, and Confluence is a great piece of Software (although there are Open Source offerings like ours that have their share of interesting features that don't exist in Confluence and allow to do very interesting stuff).

But here you are comparing Open Source software with unsupported free software. There is much more to Open Source than being free and being necessarly run by geeks having fun on their code.

There are now many companies, including in the Wiki space with professionals working on Open Source software. And this Open Source software is supported. And sometimes it's even supported with real people and in your own language in countries where closed source companies don't even have a representative.

Open Source is not about being free, it's about being Open. It's about sharing the Intellectual property. For companies building Open Source software it's about accepting more openness and sharing the property of the code itself with clients.

Now as you probably think, most customers are interested in short term benefits (cost vs value of the solution), and you are right. However if you think a little further (and we have good examples of that), on the long run the lack of competition introduced by the property being fully owned by the Software company creates higher costs and lack of innovation. So companies need to balance short term interest (the best solution at the lowest price NOW) versus long term interest (evolution of the solution, of it's cost, future innovation of this solution). This is why more and more companies are looking at Open Source solutions and this is why software companies are more and more adopting the Open Source model.

Now you are welcome to come and see us and see what we are doing to bring short term and long term value to customers with an Open Source solution. Then you can also propose a solution that has local support in the country in the customer's language.

Hello Ludovic :-)

I think the typical situation here is to deal with people who push forwards to install "some open source wiki", that is, to deploy "some kind of free code one can find on the internet", with no consideration to the managerial and organisational effort and evolutions that theorically should have been at the very root of the decision that such a tool will be installed.

"Mediawiki will be chosen because it is mainstream (because Wikipedia is mainstream)", says top-managers. And spec sheets may just confort this line. But sometimes in such a case no one who takes the decision has even *used* a wiki among the staff, and will not ponder correctly, or even at all, the usability factor. Actually the installed tool may just die with no one willing to use it. I know some examples of that, even in environments filled with dozens of geeks, that is, supposedly non-reluctant to use not-so-good tools (I think the exact opposite of geeks, blame me).

On the other hand I know many commercial -- and expensive -- collaboration tools that suck more than all the open source wikis I've met. But that in fact is no good news for equivalent tools in the open source field. This should only lead to other criteria than just functional specs.

So the point to make is in my opinion: since the wiki's adoption is so important for this KM project to succeed, how can the company ensure it? And fast?

Hi Ludo, what if I say that each of the few times I've mentioned Confluence on this blog, someone from XWiki immediately left a comment here, isn't that typical as well? ;-)

If my comment is typical of something, it's about being well grounded in reality, i.e. designing and delivering enterprise wikis to paying customers! I've done two installations of Confluence already and there are more coming. None of those projects would have worked out with MediaWiki, although it was initially mentioned by those clients. What's that if not a good reality check?

You seem to ignore (or forget) that the commercial license of Confluence gives you full access, with modification rights, to its source code. That the license gives you perpetual usage rights (speak of short term!). That Atlassian support is stellar, and complemented by people like me in just any country. That most of its plugins are free and distributed with open-source licenses... While I can see where you're going, you've raised the wrong reasons for this particular competitor.

Now I'm not married with any vendor, I just pick the good ones. So when you'll be able to talk about XWiki without complaining about competition, and understand that your product isn't the absolute logical choice in every situation, we'll talk. Just convince me, but with better arguments, please. :-)

Francois,

You are incorrect about someone from XWiki leaving a comment here. It's right it's not the first time you bash Open Source Wikis versus Confluence (note: Erwan is not from XWiki although he used to be and he is an XWiki user at his company).

Now if XWiki people are commenting on your blog, it's positive isn't it ?

If I commented here it's not because you say Confluence is good and that I complain about competition. If you read my post you'll see that I said that Confluence is a good software. I commented because you assimilate Open Source to "unsupported" and I think it's important to say that this is unrelated and that you can have support with Open Source. If you had made this comment about Wordpress I would probably have reacted also.

I have no interest here to bash Atlassian's model and I'm not. Atlassian is a great company and there is no doubt about that.

Now "Shared Source" which is what Atlassian does is not "Open Source". You don't have guarantees on future licence prices. You don't have an ecosystem which allows competition to build a different version of Confluence and resell it if Atlassian fails to make the right product in the future. Open Source gives the right to users to be in charge. Of course the "right" does not mean the "power" because being in charge can be very costly.

I'm not saying it's "wrong" to be closed source, I'm not "opposing" Open Source and Close Source. There is no problem with deciding to sell Close Source. There are great software coming out as Close Source.

However please don't undermine Open Source as a model and assimilate all Open Source software to "unsupported". Open Source business do work and create great software and innovation AND it's open. Based on the Open Source model there are very interesting things that can happen (thing alternate solution build on top of it. We are doing that at XWiki with Netcipia, Chronopolys, Nearbee, Curriki, etc..).

I agree with you however about what you say about MediaWiki being mainstream and being thought of in the first place and that without support and "professional" backing this is an issue.

Concerning our product not being the absolute logical choice in every situation, I don't have a problem with that. But you are welcome to consult us when you have clients looking for Wikis. I haven't seen you do that. And maybe we can show you benefits in XWiki that you don't see with the current level of information that you have.

Ludo, where exactly did I "bash Open Source Wikis"? Unless you imply that MediaWiki represents all open source wikis, I never made that generalization. Also, where exactly did I say all the nasty things you make me say about open source??? I've been and I'm still using a lot of open source stuff every day, some of which are perfectly supported. You wrongly generalize about my reaction to some FOSS fanatic who thinks that Open Source Is The Only Way and Commercial Is Evil!

Your argument about price and competition doesn't hold with me. The two recent price changes I've seen are: Atlassian cutting their maintenance fees by half (!) and Jives Software revising their price list for Clearspace to more reasonable prices (in just weeks, but their initial price list was a rip off). See, competition does exist and does drive prices down, you can't sell crappy wikis at rip off prices like CMS were selling ten years ago. Still, my point is that when the costs of custom developments, support and maintenance are higher with open source software than buying a commercial product that can be used off the shelf (case in point: MediaWiki vs Confluence), then it's in the client's interest to go for the commercial product. And PLEASE don't try to make me sound like I'm generalizing here, each client has different needs and the hundreds of wiki software around have different uses.

Lastly, I've evaluated XWiki, twice, and it's still on my watch list. I do a lot of product evaluations without necessarily consulting their vendors (although I often ask their clients, users, my peers...). Now if my "current level of information" isn't good enough with what's publicly available on your site and visible in your product, I'd be happy to get an update :-).

Oh, I see: "the available open source options are simply not up to par with the commercial offerings". OK, "most open source options" would have been more appropriate, I take that back since it's not what I meant to say. As an example, I've found cases where a simple wiki like DokuWiki is just perfect and other options would be overkill. Once again, and I've written that many times, it's (I'm) wrong to generalize... :-)

Francois,

"Most open source software (and MediaWiki in particular) absolutely suck at end-user features such as a really working WYSIWYG editor (no, wiki syntax isn't any "simpler" for casual users than HTML, it's just Fucking Ugly Code™ for them), or usability. And let's compare the complexity and cost of maintaining a bunch of "free" spaghetti code (bricolage!) assembled by a geek hobbyist to a product that comes with enterprise support and can be managed by non developers, etc., etc."

We are supposed to understand that "free spaghetti code assembled by a geek hobbyist" does not apply to "Most open source software" but only to "MediaWiki".

And that when saying "product that comes with enterprise support" does not mean that you think that open source software are not "products" and that they don't come with "enterprise support", but that it only applies to "MediaWiki".

"Sorry, but for entreprise wikis, the available open source options are simply not up to par with the commercial offerings. I don't take a cut in selling licenses, and my clients perfectly understand that it's in their interest to pay a license if it's less costly for them than customizing a "free" software that they won't be able to maintain without endlessly paying a developer."

This also does not mean that "Open Source" is "free" and you always needs to "pay a developer to maintain'. Again it only applies to "MediaWiki" in your mind ?

Maybe I'm paranoiac, but for me you do "generalize" in the way you write about MediaWiki vs Confluence.

BTW, it's interesting because MediaWiki is not even supposed to be an enterprise Wiki but has been build for WikiPedia type of usage. So saying that "for entreprise wikis, the available open source options are simply not up to par with the commercial offerings" and using MediaWiki as the reference does not seem right to me. What about TWiki (check the recently launched http://www.twiki.net), SocialText and XWiki.

Anyway, I think we both understand each other's point here. I do agree with you that "Open Source Only" is not an argument.

You are welcome to visit us and we'll show you what we are up to. We also welcome to your feeback and will be happy to improve our software based on that.

Ludo, please re-read Martin's comment and frame my reaction according to his.

mensuelles Archives

Recent Entries

  • Steve Jobs

    "Remembering that I’ll be dead soon is the most important tool I’ve ever encountered to help me make the big choices in life. Because...

  • Your privacy on MOTOBLUR by Motorola

    After the Nokia Ovi Store carelessness, it's now Motorola who's allowing strangers to get access to your private information on their MOTOBLUR portal. Exactly like...

  • How to resume a broken ADC download

    (I'm documenting this trick for myself to remember, but it can be useful for others…) Apple, on its Apple Developer Connection site, has a bad...

  • WTF is this ‘myEventWatcherDiv’ doing in my web?

    All of a sudden I started to find the following line in most of the web pages I was browsing, including ones I made where...

  • Your privacy on Nokia Ovi Store

    My friend Adam Greenfield recently complained about the over-engineering culture at Nokia: I was given an NFC phone, and told to tap it against the...