Measuring the Size of a Developer's Community?

Travelr9 asks: "I am engaged in a project where upper management is deciding whether to use Linux or Palm OS on a new device. Leaving aside the technical merits of each [in the context of this decision, either could work well] a key question is the size of the developer communities (for both OS and applications) of Linux vs. Palm. I have searched for info on this topic, and have come up blank aside from vague assertions. Is anyone capturing real numbers and stats? This also brings up an interesting conceptual question -- how do you measure the size, quality, impact, etc., of a developer community? Number of bodies isn't enough. Number of apps? Number of lines of code? Frequency of major releases in core application or platform categories? Can you measure a concept like 'quality of developer community' usefully?"

Re:I have used PalmOS devices since the Pilot 1000

[the_other_one]

I'm browsing /. on linux with a phoenix browser. It looks somewhat graphical to me.

If it's not graphical and really character based then can someone tell me what the ascii code is for that character that looks like a Borg head. It's kind of cool looking.

-1 Troll/Flamebait on the MQR standard

[MarkusQ]

I wasn't sure until I read some of your other posts. You seem to be rather good at the "I might not be pulling your chain..." trolls.

I suppose everyone has to have a hobby.

-- MarkusQ

P.S. Have you considered stamp collecting as a less anoying alternative?

Re:-1 Troll/Flamebait on the MQR standard

[Twirlip of the Mists]

P.S. Have you considered stamp collecting as a less anoying alternative?

Speak for yourself, man. Philately fucking pisses me off.

Twirlip (because too much karma is boring)

This raises the question...

[karmavore]

Does size matter?

The development community could be fairly large for a platform that has been around for a long time. However, this platform may be about to be superseded by a new upstart with a smaller yet growing community

And the answer, as always,

[devphil]

It's not the size of your community, it's what you do with it.

For example: /. has a huge community. Yet, every time a matter of social importance comes up, most /.ers expend no more energy than it takes to click "+1,Bandwagon" while sipping some Dew. As as a result, no social matters are affected by the /. community in the way that /. would like.

In the same vein, a developer community could be large, but if their actions are poorly planned, poorly executed, or (following the /. tradition) nonexistant, then it doesn't matter how many of them there are.

SourceForge Stats?

[Max Romantschuk]

Maybe SourceForge's [sourceforge.net] stats could help in this case... Not the whole truth, but something still.

Search for Resumes

[crisco]

I had the question of Linux vs Windows CE developer community size come up last year. I narrowed down a few different search terms for resume buzzwords and hit up Google and assorted other search engines. I also hit up some online job sites like Monster.com to see what was in demand and how many resumes were available.

By no means will you come up with accurate results but you should get a decent idea of the relative sizes of the two development communities. Bonus points for weeding out the people who have simply used either platform.

FWIW+IIRC, Linux weighed in at 9x the WinCE community.

Development tools

[Jerph]

Perhaps an easier thing to judge would be the quantity and quality of development tools to facilitate the community. I for one liked the idea of using Visual Basic on my WinCE device because it's so easy to create quick and dirty applications to meet a particular need. I never found such an easy tool for palm.

typo

[Twirlip of the Mists]

How come nobody else has pointed out the glaring typo in the headline of this article? Cliff wrote "community" when he obviously meant "penis."

Twirlip (because too much karma is boring)

- or -

[handsomepete]

You can tell the size of a developer's community by the kind of car he drives.

Linux - not just the "PDA developers" community

[jki]

a key question is the size of the developer communities (for both OS and applications) of Linux vs. Palm

I assume from the context that the new device you mentioned is PDA-like? I think that what you should not forget in the case of Linux is that you do not only have to look at the Linux PDA developers community. A big percent of what is done by the "desktop linux" developer's community can be utilized easily also in PDA -like environment. I can show you no hard facts, but I believe this is a big difference in favor of Linux. I guess, you should try to see your device and the applications & other software you fancy running on it taking this aspect into account as well.

If you're really serious

[foniksonik]

So what other industries have done for decades and hire a MARKETING firm. Now I know that the term marketing is loaded with negativity these days, especially in a community such as /. BUT when you get down to it that is what you are looking for, marketing data, your market being developers.

I don't know what data is already available from current sources that a good marketing firm could use to do analysis but they may find it as an enticing agenda and do the leg work for free, only charging you for the analysis (which of course is what you need to tell them when you propose this).

Market analysis is simply crunching numbers in a statistical manner with datasets representing your target audience or constituency, yours being developers. I can't recommend a firm but you should definitely find one with what they refer to as "Domain Experience" in technology and specifically software.

Good luck.

Palm

[droyad]

If your going for handhelds, palm is the way to go cause it's stable and standardized and entrenched. Linux on the handheld is newer and can come in many varieties. The palm OS is the same where ever you go.

Measuring the effective size of the community.

[Anonymous Coward]

It seems to me that the best measure would be to determine how many usable apps there are for devices like yours in each camp.

www.palmgear.com and www.freewarepalm.com are good places to find collections of Palm software. Bear in mind that maybe ninety percent of this is either poorly implemented, too specialized, small hacks, or buggy.

The ten percent or so of this software that is useful to a significant number of "normal" people would be the measure of the effectiveness of the Palm development base.

You'll have to use your own resources to determine how much software is available/usable for a small Linux platform.

PalmOS is really poor.

[JMZero]

When the first versions were written they didn't really plan for the future. As a result, their API is a pile of kludges. Avoid.

googlefight?

[ameoba]

There's only one obvious way to measure the size of a developer community; ask google.

Linux v. PalmOS [googlefight.com]
...but just because one is more widely used doesn't mean that it's liked [googlefight.com]

By height

[sharkey]

Ty: I never keep track of the number of downloads, Judge.
Smails: Well, how do you measure yourself against other developers then?