Former MS Employees Explore OSS

Roberto Andressi writes "A few former Microsoft employees have launched a Web site that evaluates open-source projects. The site is intended as a way for first-time OSS users to 'get their feet wet' before diving into the large community of open source projects out there. The site, Ohloh, will provide background information on a prospective project. The folks behind the site even plan to include a lexicon of terms for very new users. " From the article: "'We collect from the infrastructure the open-source community uses to develop the software,' Ohloh co-founder and CEO Scott Collison told CNET News.com. 'It also serves as an open-source directory. You can find open-source projects and compare them, and gradually find one that's right for you.' The site could appeal to developers who are frustrated by the number of open-source projects that lack clear explanations. Ohloh also seeks to help developers make a build vs. buy decision by offering code analysis, said Collison, who along with co-founder Jason Allen, previously worked at Microsoft."

Comments

Bill got em...

[XenoPhage]

Thanks slashdot.. Here I try to visit the site and see what it's all about, but it's obvious that Microsoft saw this news before me and has already squashed the site.. What a way to start a Monday...

Re:Bill got em...

[pedalman]

Either that, or these are the former MS troops who tried running Hotmail on NT.

Re:Bill got em...

[0232793]

Timid about diving into the oceanic waters of open-source projects?

A few former Microsoft employees have launched a Web site that evaluates open-source projects. The site, called Ohloh, is not a reviews site, but instead a directory of open-source software, its co-founders said.
Advertisement

"We collect from the infrastructure the open-source community uses to develop the software," Ohloh co-founder and CEO Scott Collison told CNET News.com. "It also serves as an open-source directory. You can find open-source projects and compare them, and gradually find one that's right for you."

The site could appeal to developers who are frustrated by the number of open-source projects that lack clear explanations. Ohloh also seeks to help developers make a build vs. buy decision by offering code analysis, said Collison, who along with co-founder Jason Allen, previously worked at Microsoft.

"A developer thinks, hey, maybe I want to develop this kind of project. Our formula takes into account what it costs to write the code, collect the requirements, write the specs, write the code, test that code and deploy it," he said.

Ohloh's database, searchable by project name or keyword, results in a list of suggested software. Each project has a profile, beginning with a brief synopsis of what the software does.

While other open-source databases offer this to some degree, many times developers are left wondering about licensing, Collison said. Accordingly, Ohloh also lists the licenses held for the open-source project, as well as a link to the full text of each license. (The name Ohloh refers to a cry of enlightenment in Buddhism and also the name of the first surfboard in Hawaii.)

In addition, the directory offers information such as when the project was started, how many developers are actively working on it, the languages it uses, links to the project's home page and a breakdown of current activities. Charts on the open-source project show how many lines of code have been removed or added and by whom. For those really unfamiliar with the open-source community, the profile even includes linked explanations of each open-source evaluation term.

Ohloh has been working on the directory for two years and plans to formally launch it next week. In the meantime, it has quietly rolled out its database in public beta. The database will remain free with an ad-based business model, but Ohloh plans to add services for companies looking for analysis and evaluation of their own in-house projects.

The company's investors include the two co-founders and former Microsoft executives Paul Maritz and Pradeep Singh. Collison said he and others at the company hit upon the idea for Ohloh while working at Microsoft.

"We worked on the communications pillar of Vista and Web services. While doing that, we met with a lot of corporate customers who were interested in open-source products and concepts but asked, 'How do I get an idea of how the software was made?'" Collison said. "We got so many questions from people who wanted more visibility into open-source that we thought it was a great place to start."

Re:Bill got em...

[Anonymous Coward]

Hahaha. Microsoft is a Microsoft does. Their site has already crashed. I'm getting HTTP 500 "The page cannot be displayed".

Why link to ZDNET Asia?

[assantisz]

This link is much faster (that is if you are in the USA, of course): Web site that evaluates open-source projects.

Re:Why link to ZDNET Asia?

[andrewman327]

Thanks for the link. The lexicon of terms will be very helpful, as true newbies can be easily overwhelmed by GNU, OSS, KDE, and the rest of the acronym storm that us geeks leave in our wake.

Re:Why link to ZDNET Asia?

[houghi]

I was overwelmded by the 500 Internal Server Error

Re:Why link to ZDNET Asia?

[elrous0]

The big problem with a lot of OSS project sites is the lack of basic information about what the software even *IS*. A lot of open source project sites feature build information, lists of patch fixes, etc. on their main page--all without ever explaining the basic info that people (esp. non-technical people) need to know. The are so into explaining the "how" of the software that they negelect to explain the "what" and "for who."

The lack of documentation on OSS projects is bad enough without having to spend an hour just trying to figure out what the software even DOES.

-Eric

Re:Why link to ZDNET Asia?

[someone1234]

If you don't know what is it, or whom is it, it isn't for you.

Re:Why link to ZDNET Asia?

[dieth]

The lack of documentation on OSS projects is bad enough without having to spend an hour just trying to figure out what the software even DOES.

Have you ever read any of the documentation? Microsoft has always had the least documentation on any of there OS's or Programs, and the worst search engine to help you find it on on microsoft.com

Re:Why link to ZDNET Asia?

[macintyred]

You're absolutely right about this. It's a pain trying to figure out which software does what. I think I read in the article that this site will include a profile of each project with a synopsis of what the software does. If so, I'll be spreading the word :)

I still have to see it to confirm that much though.

Re:Why link to ZDNET Asia?

[truthsearch]

It's ironic (considering this new site) that Microsoft has plenty to say about what their software is but lack in useful documentation. It's literally the opposite of OSS. Microsoft provides pleny of marketing to make sure everyone knows what they offer, yet for the technical they offer little help.

Re:Why link to ZDNET Asia?

[macintyred]

I don't know, sometimes you just have to try a few new things. Every once in a while you get lucky!

Re:Why link to ZDNET Asia?

[baadger]

That could be because most normal non-technical people never need or do visit the open source project websites (except maybe mozilla.org). Most people goto software homepages to get the software but in the Linux world, and let's face it thats where most open source software is 'sold', everyone get's it through their distro's repository.

Re:Why link to ZDNET Asia?

[gfxguy]

That's not really true. I use sourceforge a lot, for instance, and use their search. That narrows it down a bit, but then you want to make sure the software is right for you. The little synopsis doesn't provide enough detail. So I go to the project's home page, and I'm at a loss to tell you what the program actually does or what platforms it runs on; all I know is it has something to do with the something that I was looking for. Sure, sourceforge can narrow it down some, but it's not always accurate - and it doesn't really let you know that the Windows version differs from the Linux version which differs from the MacOS version...

Sure, you can usually find out what you need to know if you do enough digging, but the home page of a project, I'd think, should be almost like a brochure. If people are really involved with the project and don't want or need to see that sort of thing, they can link directly to the nightly builds or release notes or something. I just don't think the home page of a project should be the release notes.

I know a lot of open source programmers have a "take it or leave it mentality", because they've done it for free on their own time and to their own specifications... and that's fine, but you'd think they'd want to give potential new users an idea of whether or not the application fills their need. Even though you don't cry if someone decides not to use your work, it's still better to have more users than few users if you want your project to have any relevence in the long run... after all, if you've worked that hard on a project and release it for free for everyone to use, it's usually because you want other people to use it.

Re:Why link to ZDNET Asia?

[ozbird]

The big problem with a lot of OSS project sites is the lack of basic information about what the software even *IS*.

Hence sites like freshmeat.net - my first port of call if I know the kind of application I want, but not which one (or even what the available options are.)

Maybe I'm getting cynical, this has the distinct smell of astroturf. How "Ohloh" can you go?

Re:Why link to ZDNET Asia?

[Pyrowolf]

I completely agree - I can't ever find anything on the MSDN site with their own search engine. Am I the only person who uses google to find documentation on Microsoft's site?

Re:Why link to ZDNET Asia?

[ozbird]

Yeah, I screwed up the link. Use this if you can't figure it out for yourself.

Re:Why link to ZDNET Asia?

[ch-chuck]

Why should they? What's the incentive? All most developers want to do is develope software and *if* it helps someone else, great, take a copy. It's not like they make more $$ for every customer who runs it. In fact, there's a big *disincentive* to inviting the clueless in - then the devs have to spend more time answering annoying newbie FAQ's over and over, troubleshoot their installation problems, etc. If the FOSS world, the more clueful users the better; the more clueless the worse. Ergo, if you don't know what it is, you probably don't want it. FOSS: you've got to *want* it - there are no marketeers out there trying to hunt down customers, ship them product and debit their account.

Re:Why link to ZDNET Asia?

[Skye16]

You make the relatively large assumption that every person using OSS software is using Linux or *BSD. I try to use OSS in almost every aspect of my life except the OS. (Just to head off any questions as to why: I'm a gamer. I play games that *I* enjoy. These games are *only* found on Windows. I'm not going to not play these games just because someone else thinks they have better alternatives. If I thought as they did, I wouldn't be using Windows. But I don't. So I do. Okay? Good.)

OSS projects don't exactly have to kill themselves giving this information, you know. Take squid, for example. It says what it is, it tells you what it supports. Straight up, in your face, nice and easy. The main page is not a simple HTML list of appended patch notes for the last 3 years. It's not too hard to make your main page be a very brief, but telling synopsis of what your software does, while still maintaining your precious bug fixes and other release notes on another page. If your goal is to make the 5 people already using your software happy, then sure, continue with what you've been doing. If you're trying to get another 5000 people to use your software, you're probably going about it all wrong.

Re:Why link to ZDNET Asia?

[buraianto]

I can't speak about most software, but .NET is pretty well documented for most things.

Re:Why link to ZDNET Asia?

[Angostura]

I take it you haven't tried using Mac OS X's pitiful built in help. MS's effort leave Apple's standing. And I don't like saying that one bit. Not only is the system slow, but the content is very, very thin.

Re:Why link to ZDNET Asia?

[greenrd]

Yes, but that's not surprising. Microsoft is all about "Developers, Developers, Developers". I don't use MS products but I'd be interested to hear what the standard of their documentation is for things other than core programming APIs.

Re:Why link to ZDNET Asia?

[denim]

There's no way to know that for sure if there's no description. I'm pointing an aspiring tech writer at that site.

Re:Why link to ZDNET Asia?

[truthsearch]

I was a .NET developer for a few years. The documentation is terrible. The sample code is no more elaborate than "Hello world" and many methods get a one-line explanation with no details of what changes in different scenarios. VBScript has better documentation. Scary.

Re:Why link to ZDNET Asia?

[kimvette]

An apple fan would tell you that you don't meed the online help. If you cannot find it in the help, it's something you don't need to know.

(I kid, I kid)
*eyes sawtooth that is going to be running Linux soon*

Re:Why link to ZDNET Asia?

[Kazoo the Clown]

Documentation is certainly the biggest weakness of Open Source IMHO. Almost every library I've seen of useful functions considers an auto-generated class tree and/or couple of demo programs as "documentation." I'd be embarrassed to release something like that, but I guess I'm in the minority...

The second weakness I'm finding out is projects that the original developers have essentially abandoned. You find forums that have recent posts by people asking questions about it but noone with any answers....

Re:Why link to ZDNET Asia?

[Anonymous Coward]

Have you ever read any of the documentation? Microsoft has always had the least documentation on any of there OS's or Programs, and the worst search engine to help you find it on on microsoft.com

Maybe you should use Unix as your benchmark, instead of Microsoft, unless you're only trying to find an example that might put the very poor state of OSS documentation into a slightly better light (assuming you're correct about Microsoft documentation -- in my experience, Windows documentation is typically better than Linux documentation, but worse than BSD or Unix, with the latter tending to be the best).

Re:Why link to ZDNET Asia?

[WuphonsReach]

VBScript has better documentation.

Maybe because VBScript is simple enough that it's easily documented?

Not sure whether that's a joke or not...

Alternately, I suspect that the documentation is better because VBScript was (at the time) going up against other scripting languages for mindshare back in the late 90s. Poor documentation might've slowed uptake of the new language.

Re:Why link to ZDNET Asia?

[Anonymous Coward]

It is the responsiblity of the software maker to market their product and provide the proper documentation, marketing materials, white papers, etc, just like you would expect of any other software companies. If there were truly any value in these projects, then it would be easy for a company to evaluate them for their needs. If a OSS project can't be bothered to provide the resources a potential clients needs, then I can't be bothered to pay them any attention either. Yes, all of that costs money, but that's what capitalism is all about right? You need to invest a little now to get back a lot later. Oh, wait "You can't own anything man" says the hippie. Well, yes I can because I'm not a filthy hippie.

This isn't something that is at all uncommon. Thousands of normal ISV's do it every day, including me. This is typical of OSS projects, "I wrote the core code, somebody else can do all the boring stuff".

I had high hopes for FOSS in the beginning, and since then it has never ceased to dissapoint.

Re:Why link to ZDNET Asia?

[Anonymous Coward]

He is refering to the link to Ohloh not zdnet dude... he'll face the same thing from any links...suspect the the can't take the slashdot traffic...

Re:Why link to ZDNET Asia?

[jesterzog]

I've found .Net API's to be quite well documented, at least for what I use them for. If I was only developing native .Net applications that never interacted with anything but .Net interfaces, it'd be quite nice. My own problems have come up when interacting with non-.Net API's, which tend to involve unmanaged memory, and I've been quite surprised how many Microsoft products still seem to rely on them given MS's push for everyone to use .Net.

For me, the Office API has often been really hard to use effectively. It's quite buggy, and the documentation seems to be sparse. There's lots of documentation-by-example for doing specific things, but little specification of what the different bits of the object model actually represent or how they're supposed to act.

Re:Why link to ZDNET Asia?

[CptPicard]

And when they do try to explain what the software is and does, it is explained in such a way that half of the text consists of abstract jargon and buzzwords. You really need to be in the loop already to even begin to understand what the supposedly helpful explanation is supposed to mean. In my experience, Apache's Jakarta projects are particularly guilty of not being able to explain their components in plain English.

Re:Why link to ZDNET Asia?

[Anonymous Coward]

The only thing scary is how fucking stupid you are. I was doing Java back in the v1.0 and 1.1 days and the documentation was the same way. VBScript has more and better docs simply because it's been around longer. Duh. You think because you see something for the first time, it is the first time. I sincerely hope you're only like 10.

A reminder for all companies

[Marcos Eliziario]

Treat your employees well. They will still be around for a long time after they leave your company (unless, of course, your company is the mafia)

Re:A reminder for all companies

[Anonymous Coward]

Or unless of course your company is "The Company"

Re:A reminder for all companies

[ObsessiveMathsFreak]

But why be nice, when you can Sue(TM)?

Uh Oh, Down Already.

[Anonymous Coward]

Slashdotted before the first 10 posts.

---!

The new controls on slashdot are horrible, you should burn!

Re:Uh Oh, Down Already.

[Anonymous Coward]

I agree, this crap is really getting annoying, about the only thing missing now is for SD to add some pony background music that you can't turn off. I mean really, a laggy floating options show hide button!? You know you're being a design asshole when half your viewers turn off javascript because of your "awesome" design ideas.

Some smart MS employees...

[Anonymous Coward]

are ralising a great business opportunity!
1. Work at MS, get to know where those smart coders steal their code from.
2.Build an index of info. already available from Sourceforge.
3. Get advertised on Slashdot.
4. In every resulting page, display dozens of shoddy MS products in Google ads.
5. Profit!

At this rate, wonder if any chairs are still left over...

Term that they should have added

[Rik Sweeney]

Slashdotted

This is what happens to a site when Slashdot links to it on their front page.

The irony of course is that no one would be able to read it when a Slashdotting occurs...

Re:Term that they should have added

[dragonman97]

It's been ages since I've seen it referenced here, but it turns out Mirrordot does have it cached!
Ohloh (front page only)

Re:Term that they should have added

[Anonymous Coward]

How can you be sure that these were ex-MS employees? Their pictures have been digitized and their voices altered so that nobody could point a finger at them...

Not to be outdone...

[AngryDill]

Larry Ellison has announced a new web site where he evaluates the benefits of PostgresQL and DB2! ;)

-a.d.-

Re:Not to be outdone...

[anomalous cohort]

I searched for http://www.mono-project.com/ on http://www.ohloh.net/ and the only relevant hit I found was Qt#. Imagine that.

Open Theft

[Doomedsnowball]

In related news, the start of a new project at Microsoft to figure out exactly how much code needs to be changed to patent OSS ideas and inline them into Vista. The lawyers say 80% of the comments need to be altered. The QA testers say 51% more bugs need to be added. And the developers say that you just need to change all the names of the variables, make pointers to everything you can, and cut-n-paste inline assembler code before the linker kicks in.

Re:Open Theft

[mac.man25]

That was modded insightful?

THIS POST IS FUNNY!

Re:Open Theft

[$RANDOMLUSER]

You forgot the part about running the docs through the remove useful information, translate to baby talk and insert cross-references to unrelated pages process.

Re:Open Theft

[M$ Mole]

Seriously...how did this get modded "Insightful"? I can understand someone modding it as "Funny", but the comment is not insightful in the least.

As Ackbar would say:

[mgblst]

"It's a trap"

Only kidding, I am sure that a Microsoft employee could turn from the darkside. (I think thats enough starware references for this week)

Is this any different to any other open source sites out there - is it especially for windows users. (if not, I am not sure of the reference to an ex-microsoft employee)

Re:As Ackbar would say:

[xmorg]

my thought exactly. They are probably plants.

Re:As Ackbar would say:

[badfish99]

It's very slow, but I have managed to look at a couple of pages on the site.

There's a line or so of information about each project (e.g. for Apache Ant it says "Apache Ant is a Java-based build tool", which is not very helpful). And then there's an estimate of the total cost of the project, based on an estimate of the number of man-years that have gone into the code, costed at $55k per man-year.

WTF is that all about? Are these people perhaps trying to suggest that open-source software is valuable, and that its developers should charge for it?

Re:As Ackbar would say:

[mrxak]

Nah. If the site was run by microsoft, it wouldn't have gone down so easy to a slashdotting.

Re:As Ackbar would say:

[rucs_hack]

"WTF is that all about? Are these people perhaps trying to suggest that open-source software is valuable, and that its developers should charge for it?"

Stage 1: Establish tht open source has a cost in real money.

Stage 2: Get Hired as consultants by microsoft to state this in press releases. .... well, you know the tune

Re:As Ackbar would say:

[Jimmy King]

"WTF is that all about? Are these people perhaps trying to suggest that open-source software is valuable, and that its developers should charge for it?" Stage 1: Establish tht open source has a cost in real money. Stage 2: Get Hired as consultants by microsoft to state this in press releases. .... well, you know the tune

I don't think that's what they're going for. I think it's a more good intentioned "this is a rough estimate of what this software is worth if you were to develop it/this is what it cost the people who are now providing it to you". Of course, I can't imagine how it's accurate... are they figuring a 40 hr work week just on that software for each person involved for however long it's been in development? Are they using some estimate based on software is X lines of code and on average a developer can write Y lines of code per day/week/month/year?

Re:As Ackbar would say:

[rucs_hack]

I have no faith in Lines of Code cost/time estimates.

Perhaps in a workplace with fixed quota's that might work, but I code at night, or when I'm in the mood, or when I've been for a nice walk, and I seem to be plenty productive enough. I doubt it could be costed easily though.

Newbies guide to opensource

[Kuku_monroe]

Gettin' your feet wet' before diving into the large community of open source. Step 1: All Open source projects have weird name like "Ohloh"

Redemption

[Rob T Firefly]

"Scott Collison!" I exclaimed. "But.. you left Microsoft!"

"Collison?" He appeared pensive. "Yes... That's what they used to call me. Scott Collison. That was my name." He smiled.

"Scott..?" I asked, confused.

"I am Collison the White," he proclaimed, with a twinkle in his eye. "And I come back to you now at the turn of the tide."

I wonder why ...

[rowama]

I'm having a hard time with this. I get the feeling this is a case of "If we (i.e., M$ employees) build it, they will come." Since ohloh has been slashdotted, I cannot tell if this is just a freshmeat knockoff (with a little sourceforge and krugle thrown in).

The site is intended as a way for first-time OSS users to 'get their feet wet' before diving into the large community of open source projects out there.

I got as far the frontpage (hehe), and I think newbies will be afraid to stick their feet in much less dive-in.

Sorry, but I may not be seeing clearly through these suspicious eyes of mine. If so, my apologies to the ohloh folks. Also, welcome and aloha.

Re:I wonder why ...

[jgmitchell]

I got as far the frontpage (hehe), and I think newbies will be afraid to stick their feet in much less dive-in.

I got a couple of pages into it, but it didn't seem like it had that much useful information on what things are and I looked at a couple of the "bigger" projects.

I'm sorry, but how does knowing how much time developers spent on the project, the estimated cost of the project and statistics such as the total number of lines of code help a newbie?

"This project has over 100,000 lines of code. It has to be awesome!"

The Oh in Ohloh

[aapold]

Well, when you leave M$, might as well go do what you want...

Re:The Oh in Ohloh

[Anonymous Coward]

So, who is loh?

New terms...

[unforkable]

The folks behind the site even plan to include a lexicon of terms for very new users. Do they mean terms like: security, robustness, community, help, freedom.....?

it's a typo

[DrSkwid]

Oh-noh!!

del.icio.us clone

[Fazed]

Looks like a market targeted version of del.icio.us to me. Which in itself is not a abd thing as the tagging metaphor works quite well for this type of application. It will be interesting to see if this project has any success.

Re:Free Software Descriptions

[Anonymous Coward]

There's a pretty good source of popular free software application descriptions on Saugus.net's Computer Knowledgebase.

SIgn me up!

[shmert]

I'm most excited about a central place that does code reveiw of open-source projects. That's really a tedious process when you're wading around in sourceforge trying to find a shared lib for your project. Usually you can tell by the level of polish applied to the project's website how organized the code will be, but I'm certain some well-engineered software gets passed over if this is your only criteria for quality. Someone designing an open-source product shouldn't need to design a flashy website to promote it.

If there were a comprehensive site I could visit that had evaluated (albeit briefly) some of these packages, that could be a big time-saver.

Re:SIgn me up!

[Aldric]

Someone designing an open-source product doesn't need a flashy website - all they need is an informative website laid out in a sane fashion. Too often, open-source projects have only a couple of lines of text on their site that doesn't come close to even explaining what the software does.

Ohloh?

[autophile]

I would have thought these former Microsofties would have named the site "Ohnoh", pronounced with an appropriate amount of dismay.

--Rob

Freshmeat

[ozmanjusri]

It's hard to tell since their server's cactus, but it looks like it's imitating http://freshmeat.net/ (Disclaimer: another part of the nefarious OSTG network). Freshmeat is;

• "searchable by project name or keyword, results in a list of suggested software."
• "has a profile, beginning with a brief synopsis of what the software does."
• "lists the licenses held for the open-source project, as well as a link to the full text of each license."
• "offers information such as when the project was started, how many developers are actively working on it"
• "the languages it uses"
• "links to the project's home page"
• "a breakdown of current activities"
• etc

and has done for donkey's years. These guys have learned MS techniques well...

Linux cache

[truthsearch]

That's true, but only because of the large array on Inktomai linux servers they use for caching. Ironic.

Re:As _Ash_ would say:

[KiloByte]

"It's a trap"

-- get an axe.

ho | Ho

[Tx]

Reverse the name to get the true meaning:

Ohloh -> ho | Ho

Ex MS employees "helping" FOSS with a project called "ho | HO" (backwards)? Do I need to polish my tinfoil hat, or is that suspicious?

Re:ho | Ho

[ichigo 2.0]

I think that tinfoil hat is quite shiny enough.

Years of work

[truthsearch]

That is interesting. But what I'd really like to know is how they determined the time. I've never seen an OSS project which publishes how many hours of work when into writing it. It must be pure speculation. Since they're targeting the OSS illiterate too many people are going to believe these made-up numbers.

Re:Years of work

[badfish99]

Looks like they're assuming about 3600-4000 lines of code per man-year, but it varies from one project to another: perhaps they apply a random fudge-factor to disguise the simplicity of their estimating process. It doesn't say whether they are counting blank lines and comments: if so, we'll have to start padding out our code to push its value up.

Of course, the whole purpose of this will be to point to the "high cost" of free software, when justifying the "lower cost" of Microsoft products.

The descriptions of the projects seem to have been copied from Freshmeat, so this site certainly does not add any useful information about them for first-time users.

Re:Years of work

[Speare]

There are plenty of source-lines-of-code value estimators out there. One which I have used out of curiosity was sloccount (http://www.dwheeler.com/sloccount/), which took various models of code-lines per man-day and extrapolated from there.

Meanwhile...

[marcello_dl]

... I decided to convert myself to Islam, and I'm looking for Jewish sites with good introductory material for my new religion.

Re:Meanwhile...

[EsbenMoseHansen]

I wonder if you would get more - or less - factual information than if you sought your information from a Islamic site.

Anyway, the site will survive or die on it's own merits. Personally, I just use wikipedia :o)

Opportunity cost

[wolf87]

I think that the most interesting piece of the story has been missed. Ohloh is looking to create a paid service that will assess proprietary software for organizations. Combined with their open-source project DB, I see them building a set of tools that would enable IT managers to assess the relative costs & risks of different solutions far more easily. I would be looking for two developments to make this a reality. First, Ohloh would need to create an estimator of risk for each project based on the available data (a 'risk score'). Second, they would need a way to estimate the cost of customizing existing solutions (open-source & commercial). If they get both of these, they could provide a standardized, risk-adjusted measure of the opportunity cost associated with each development/deployment option. Really hope they are heading this way.

Re:Opportunity cost

[P3NIS_CLEAVER]

.....and the 'uninteresting' part of the story made the headlines. There are probably 100,000+ ex Microsoft employees. Who seriously gives a shit and what does this have to do with anything?

Re:Opportunity cost

[mrchaotica]

I just hope they aren't doing it for the nefarious purpose of inventing "data" to claim that Free Software is more expensive.

In fact, it's really too bad: this would be a good thing if it was certain that they'd make a fair and impartial comparison, but the fact that they're ex-Microsoft automatically puts them under enough suspicion to ruin the usefulness of their service.

Re:Opportunity cost

[Anonymous Coward]

It is simple
microsoft + slashdot = flamefest

Some people out there just hate microsoft 'just because'. Oh they can have their reasons. However most of the time it is just plain hatred for the sake of looking cool.

Just follow the MS plan and you will like all of their software
version 1.0 do not use.
version 2.0 hm neat idea do not use.
version 3.0 beta usable and in someways better than what is out there. Use at own risk.
version 4.0 out of beta still a bit buggy. Getting interesting however, much like all the other kinds of projects out there. Usable tad flakey.
version 5.0 must have. At this point they have got it together and usually beat everone in features.
version 6.0 rehash of 5.0 fixing most of the bugs adding all the features that didnt make it into 5.0. Again must have.
version 6.0.1+ bug fixes, bloat. They sit around adding useless features that few will use. Get for bug fixes.

Just be careful sometimes they start in the 'middle' with their version numbers. You just need to know what the real version number is. Also always wait for sp1 before production use.

Xbox 360 is basically a 3.0. Dreamcast then xbox then xbox360. The next one should be 'interesting'. The next windows is about version 6.5-7.0 it will be 'buy it to get bug fixes' not because you need it.

Re:Opportunity cost

[Anonymous Coward]

The next windows is about version 6.5-7.0 it will be 'buy it to get bug fixes' not because you need it.

Actually, the next version of NT is approximately version 4. The pre-NT DOS/Windows OS (a different OS that just shared the same name, and in version 4.0 and above, implemented the same API) ceased being developed after the release of Windows Me (version 4.9).

The versions of the NT operating systems are roughly:

v0.00: NT development started (1988)
v1.00: NT 3.1 = Windows NT 3.1 (1993)
v1.50: NT 3.5 = Windows NT 3.5 (1994)
v1.51: NT 3.51 = Windows NT 3.51 (1995)
v2.00: NT 4.0 = Windows NT 4.0 (1996)
v3.00: NT 5.0 = Windows 2000 (2000)
v3.10: NT 5.1 = Windows XP (2001)
v3.20: NT 5.2 = Windows 2003 (2003)
v3.20: NT 5.2 = Windows XP/64 (2005)
v3.20: NT 5.2 = Windows 2003/64 (2005)
v4.00: NT 6.0 = Windows Vista (not yet released)

Me Too... That is why I patented that method...

[Dareth]

... so um work hard slaves! And don't bitch when I shut you down until you fork over the ran...um royalty!

Coral Cache

[Anonymous Coward]

This sucker's been slashdotted.

In case there's still somebody on Slashdot who doesn't know how to get the Coral Cache: http://www.ohloh.net.nyud.net:8080/

Many important projects missing.

[need_change]

I couldn't find OSS staples of Java development: Tomcat, Hibernate, Rhino, Spring, Xerces, Xalan, JSON, Jakarta Commons, etc. I wonder by what criteria projects were initially included.

Re:Many important projects missing.

[Medievalist]

I couldn't find OSS staples of Java development: Tomcat, Hibernate, Rhino, Spring, Xerces, Xalan, JSON, Jakarta Commons, etc. I wonder by what criteria projects were initially included.

It's worse than that, they don't even list samba...

Similar to...

[Anonymous Coward]

As someone who had to choose an open source CMS package recently, I found this site much easier to use: http://www.opensourcecms.com/

where is the evil?

[WeAreAllDoomed]

i can *smell* the evil, but i can't see it yet.

why are they using lighthttpd?

[Anonymous Coward]

Maybe they could have handled the load if they were using apache and not lighthttpd. From netcraft:

http://toolbar.netcraft.com/site_report?url=http:/ /www.ohloh.net

Re:why are they using lighthttpd?

[Anonymous Coward]

Alas, Apache hasn't any magical power to allow a server (or group of servers) to service more requests than the hardware is capable of doing.

Propietary cost

[guabah]

Did you noticed how they estimate what it would cost to develop this software if it was propietary?

Because I don't remember seeing that anywhere else

Re:Propietary cost

[guabah]

Sorry for the double post but...

According to them, the Linux kernel is cost 78million to develop

Re:Propietary cost

[Xtifr]

What kind of QA costs are they assuming? If they're basing their estimates on MS's QA costs, they may be coming in as much as an order of magnitude low! :)

It doesn't seem to say anything about how they do their estimation. An accurate estimate would have to take into account (among other things) bugs per LOC, a number which is VERY expensive to reduce and hard to measure (especially from the outside).

Re:Propietary cost

[guabah]

I think they're speaking of initial time to develop.

Re:Propietary cost

[Xtifr]

QA costs are part of the initial costs to develop. And one of the most widely variable parts. Higher quality code costs more to create. That's why I'm curious how (if at all) they're estimating the quality costs.

Re:Propietary cost

[init100]

According to them, the Linux kernel is cost 78million to develop

And according to them, the Linux kernel project started one year ago.

Overheard at Ohloh...

[GC]

CEO: Good news guys, the web devs have completed the search facility and our website will go live next week.
Lowly Ex-MS Employee: So what domain name have we chosen?
CEO: Good question, I'll let our CTO answer that one.
CTO: ahem, well as some of you know, we registered ohloh.org, ohloh.com & ohloh.net through "Proxy" registrars before the final decision was made. In the conference call with Bill we had last week we made our decision.
Lowly Ex-MS Employee: and?
CTO: Well the decision was made that "ohloh.org" would be ditched... it's simply to difficult to masquerade as a .org organisation - after all, we clearly want to profit from OSS some way or another. The .com address, well we decided that it should re-direct to the .net address. For Marketing purposes we will use the .net address as it is sufficiently vague as whether it is a profiteering company of open source ideals or an open source organisation without profit aspirations.
CEO: I think you'll all agree that this is an ingenious idea.

ohloh.net whois entry
ohloh.com whois entry
ohloh.org whois entry

Will this be useful?

[thetoastman]

I did a simple search on the site to see how it performed. The results are interesting, but I'm not so sure about the utility of this system.

Search entry

First of all, the searches are very simple. I can only enter a topic to search for. I cannot restrict the search by language, environment, license, or platform. Grouping of search terms does not seem to be possible. Searching for project management software returned results for volume and database management as well as project management.

Search Ordering

There does not seem to be any ordering in the results. The ordering does not seem to be alphabetical, by activity, by language, or by relevance. When 200 entries are returned at approximately 24 per page, some ordering would be nice. Only a dedicated searcher would move past the first three or four pages.

Individual Results

There are a lot of problems with the entry for a particular software package. Here are a few (in no particular order.

• Advertisements are mixed with content and have no distinctive styling
• Start date is not given - only a statement that the project was started n years ago
• Code costs are interesting, and how they computed it is anyone's guess
• Developer activity graph means that most people delete and add the same number of lines?
• Code contribution graph means that most people delete and add same amount of code?
• There is no mention concerning the last update of the package
• The home page for the project is at the bottom of the web page in small type
• Factoids are inconsistent. A potential licensing conflict in one project is not mentioned in another project with the same licensing issue.

Comparison

Labeling software as abandoned is not positive or informative. For example, one software project was labeled abandoned even though the developer considered the project ready for production.

Here are some other issues.

• The summary information is not sorted in any meaningful way.
• Advertisements are interspersed with content but have no clearly differentiating style
• The graphs are confusing.

Summary

The site is still beta. However, there are a lot of issues that make this site less than useful. These issues are.

• Poor search definition
• Unordered results
• Undifferentiated page layout which makes key information difficult to find
• Home page of a project at the bottom of an entry
• Comparison metrics are not useful

Gee.

[Anonymous Coward]

Some bitbrain intern looked at the word linux, or bitbrain billionaire bill gates was suckin off bitbrain linus' "effect" cock AGAIN? rofl.

Interesting math in their metrics

[lennier]

http://www.ohloh.net/opensource/software/coppermin e

Project Started 2 years ago
Active Developers 2
Codebase 189,002 LOC
Effort (est.) 49 Man Years

2 x 2 = 49. Hmmm.

Re:Interesting math in their metrics

[init100]

Right, and the Linux kernel project started one year ago.

MS & Open Source fit well

[aoporto]

It's a good thing that stakeholders of MS and open source are finding how they can coexist. In addition to site like Ohloh.net and CodePlex, there are more emerging open source applications for the .NET platform. To avoid IP issues, a BSD license may be preferred by the MS crowd, and is featured in applications such as DotNetNuke and ListRing. The idea that it is MS versus open source as a whole is simply not valid as more for-profit organizations get into the mix. It is simply a matter of finding which approaches to open source fit best.

two other sites already

[meganc]

I think some folks are already doing this, and not just Freshmeat either as some of the previous postings have said. (1) See Alex Boxworth's Swik, which is a wiki about Open Source software projects. (2) Also see Business Readiness Rating which is a framework for evaluating open source software.

Open source sure can kick a gift in the mouth

[Finkbug]

"Is this any different to any other open source sites out there - is it especially for windows users. (if not, I am not sure of the reference to an ex-microsoft employee)"

You'd prefer that info wasn't upfront, to further encourage the Evil Empire and tinfoil cap crowds? And if IS for windows users, ain't it a potentional switch campaign for open source proponents? Jeez.

Parisans are so tiresome. Angband rulz Nethack!!!!!!!!!