use Perl

We had a BOF ("Birds Of a Feather") for marketing Perl yesterday. Dave Cross organized it and I kicked things off. My agenda didn't quite go as planned. I started by laying out the goals:

1. No solutions were to be proposed. I stressed this several times and people apologized and then presented their solutions.
2. We needed to define a clear goal of what we were hoping to achieve.
3. Once a goal is defined, determine what information we needed to achieve that goal.
4. When we know the information, we could then start work on understanding how to collect this information.

The BOF didn't follow that. I didn't think we'd get all of these concerns addressed, but when we had 51 people showing up and pulling things into different directions, it became herding cats. I tried unsuccessfully to pull things back on track, but everyone was insisting on explaining the problems they had with Perl and the solutions they saw for the problems. After cutting a few people short to ensure that others got a chance to talk (frustrating because some people who went on at length had excellent lists of issues that they wanted to address), I went with the flow and steered things from there. Here's what came out of it.

We couldn't agree on a clear goal, so I just pulled rank and said "our goal is 'bringing the excitement back to Perl'" and we moved on from there. Then one person (I didn't catch his name, sorry), asked the annoyingly obvious question: how do we measure success? We have several ideas here, but we need more. Job postings, analyzing server logs from perldoc.perl.org (a lovely site, by the way) and other Perl sites, tracking articles and similar things. We don't yet know the most effective things that will be needed here, but we know we need this.

Richard Dice suggested that when companies are worked with (he had a lot of say about the selection of companies), we need to have a neutral way of getting information. Imagine a matrix with the vertical axis listing value drivers ("availability of programmers", "speed of development", "commercial support" and so on) and the horizontal axis listing various technologies and have them "score" their perceptions of these technologies. I've no idea if this is the way we'll go (yes, we know we need to talk to more than just companies), but it's much better than what we previously considered.

There was also general -- not universal -- agreement that we need to have market research and analysis professionals involved. How we pay for this is unknown at this time, but Gabor and others had plenty of ideas on how to make this happen.

Edmund von der Burg, the brains behind send-a-newbie.com, surveyed the newbies for their attitudes. It's not a comprehensive or formal survey, but that information was illuminating nonetheless.

Some of the various issues which were raised (no endorsement, just mentioning issues):

• No one hears about important Perl projects. Everyone knows about Rails, Django, WordPress, etc. Perl doesn't have a lot of visibility outside the community.
• Perl sites are ugly. If you here Perl is dead, our community will bear that out. Compare the first Google hit for Ruby and the first Google hit for Perl. It's embarrassing.
• TIMTOWTDI is a problem. Companies are finding it makes system administration a nightmare and newcomers complain about being lost in a sea of choices.
• People outside of Perl think that Perl 5 is waiting to die and be replaced by Perl 6.
• Programmers trying to learn Perl are having trouble finding resources online. The first Google hit for "Perl tutorial" is a tutorial for Perl 4.

There was quite a bit of information that I learned which I didn't previously know, but perhaps one of the most fascinating things was this: in many countries, Perl does not have a negative reputation because no one's heard of Perl. If that's not an opportunity, I don't know what is. That might be beyond our scope (I really don't know and it would be hard to quantify), but I highly recommend people think about how we could change that.

Obviously there's a lot more than that and we have more work to do, but this is a good start. Some potentials can't be discussed yet (there are some companies which might be willing to provide advice/money/assistance) and others aren't finalized, but we're starting to get things on track.