June 04, 2005

It's a funny old world...

Almost five years ago (five years next month) Perl 6 got set in motion, at whichever TPC it was. (4? 5? I don't remember) We had a whole bunch of people volunteer to handle bits of the project -- I took on the dev lead hat. (At this point I think, of the original volunteers, only Larry remains) A couple of months ago I gave up the hat, and now I'm giving up parrot development altogether. Chalk up another developer driven away.

I may spend some time writing up explanations of why things were done the way they were over the next few months, along with designs for some of the systems that never got (and likely never will get) implemented. I'll probably let the blog peter out to nothing after that. I'll leave it up since there are links into it, but if I feel the urge to blog I'll start up something fresh elsewhere, since it likely won't be particularly technical and it'll forestall the whining about me writing about things nobody cares about.

And no, my post-mortem on things won't be public, though I am going to put one together. It's always a good idea to look back on why you've bailed on a project.

Posted by Dan at June 4, 2005 09:52 AM | TrackBack (0)

We may be few and far between, and not particularly vocal, but there are some of us that have enjoyed the squawks. Looking forward to a recap of some of the design decisions. A post-mortem might be interesting in a Jerry Springer sort of way. But it mostly sounds like you need to recharge yourself. Be selfish for a while. And try to put some fun back into whatever you do.

Posted by: Chris Rathman at June 4, 2005 09:42 AM

I also really enjoyed this blog, particularly it's technical content, and I'm very much looking forward to the recaps. But I understand the need to step back - and it makes sense that the stepping back includes the blog. But I'll miss it none the less.

Posted by: Nathan McFarland at June 4, 2005 11:03 AM

Far too long have I listened to the backroom grumbling and dishing and cynical realities of this white elephant from far, far too many people who haven't the balls to say it out loud and instead let it fester. I've tried in earnest to get my own cynical and detached old man to go public with his complaints as he, too, has largely walked away for many of the same reasons you have. I understand a level of weariness that leaves little caring for the scorn of the blasphemer, but the cult of perl could use a reality check if you're willing to serve it.

Posted by: hfb at June 4, 2005 01:12 PM

What'd be the point of dishing the dirt? Really, honestly, what point? I didn't when I was running stuff because it wasn't worth it -- there was a lot more downside than up. And now? Wouldn't change anything. The people who know what crap's going on would still know, and the people who don't either wouldn't care or would scream, and the screams are part and parcel of what drove me off.

The only thing that'd truly happen is I'd come off as a bitter and petty. I'll pass right now.

Posted by: Dan at June 4, 2005 06:13 PM

Whatever the issues that have been tearing apart the Perl community for the past few years are, I've seen them drive away some of the coolest people I know from the Perl community. I can tick off a list of folks that will no longer have anything to do with Perl development, and it saddens me to see this affliction strike you down as well. I hope that someday it is resolved at the source by whoever understands what's going on, as it's not fair to ask those leaving to try one last time to make things different. That hope dims a little more today. You'll be missed.

Posted by: Richard Soderberg at June 4, 2005 07:06 PM

Dishing and actually just saying something, anything, about it are two different things. Even Autrijus has started to notice things are very not right in the woodshed, but you have to get close, real close, to get to the really fcuked up parts that few see and noone talks about. I'm tired of being the goto guy for the newly cynical realists who have finally figured out that I'm not a bitch afterall, rather a canary who likes to drink, smoke and torment small boys gonzo for perl. It's just sad to see it end with a whimper and a few god complexes driving off everyone worth keeping.

Posted by: hfb at June 5, 2005 02:35 AM

Although perl isn't my favourite language, I've always found your writing interesting and enjoyable and I've often directed people, by way of explanation, towards your "What the heck is..." entries. Hope you get fired up about something new. :-)

Posted by: Andrew Birkett at June 5, 2005 02:25 PM

I hope you put your designs up, at minimum-it would be a shame to waste all that design work. Do you think perl can be saved? It was the first language that I really liked, and I hope that it can survive, but I worry, if the community around it is so toxic as to drive you off.

Posted by: matt knox at June 5, 2005 10:11 PM

Dan, when several major GCC developers got fed up, they forked. Could you gather some like-minded folks and do the same? Or would there be too much excrement flying your way?

Posted by: Chris at June 5, 2005 11:43 PM

It's really a pity what's happening here. Although I did not use Perl for more than two or three years now (I'm using Ruby now), I've watched the development of Perl6 and especially of Parrot very closely. Hence, I've read both your squawks and your books with great pleasure. Perl6 doesn't matter much to me anymore (there are better alternatives readily available), but I hope to read you again somewhere soon.

Posted by: Maik Schmidt at June 5, 2005 11:45 PM

Your blog has been instrumental in interesting people in Parrot and all the difficult concepts like continuation. You know that.

A substantial part of the Perl6 undertaking is getting people to speed.
It takes time. Things are slow to get in mention.
Remember Larry's joke in one of the Apocalypse.
The acceleroraptor will eventually get to the velociraptor.


Posted by: cognominal at June 6, 2005 10:15 AM

Dan, there are younger guys (like me) who aren't as deep into Perl development as you are. A lot of them will listen to you. So if you have something to say that'll help them make the right decisions, please say it.

Posted by: Brent Dax at June 6, 2005 12:32 PM

well, I loved the technical stuff on the blog, thanks for it even if you abandon it

Posted by: riffraff at June 6, 2005 03:28 PM

At some point I am going to come clean -- I've been convinced that it's ultimately beneficial, and now that I've got no particular interest in the process (running the show means you hold your tongue a lot. Anyone who thinks there's no politics involved in their project has nobody else involved in their project) some honesty is in order, if for no other reason than maybe someone else'll avoid some of the mistakes I made, and if not avoid 'em at least recognize when things are going bad and fix the problems.

I'm not abandoning parrot as a chunk of software. We're still using it at $WORK, and I'm not inclined to rewrite the compiler to emit x86 object files. I'm just another person who's dropped the process and left.

Posted by: Dan at June 6, 2005 05:44 PM

I believe the reason is that They Can Not Code. After five years of talking, that's about 1800 days, there are no interesting result of the Parrot. How can you keep your moral up with 1800 days of nothing?

Perl, Python were put into shape in maybe less than 3 months. Yes, a much simple one. And kept involve to the complex mass (mess) of today in 10 years.

Parrot was not in any workable shape in 5 years, or 60 months. And it will not be able to involve since people don't know what it will look like.

Pugs on the other hand, produced something that people can see from day one. That's how it attract developers. Developers attracted to pure talk/vapourware will not last long.

Posted by: ac at June 6, 2005 08:24 PM

Mmm, the squawking of the sidelines. Wondered how long that'd take.

That could be answered by noting how far Parrot's gotten with so little actual time put in (it's got perhaps ten man-years, on a project that's been going for less than four and a half years, and a good third of those were spent slapping yet another coat of paint onto the bikeshed, and another third in throwaway code that's been thrown away). Or it could be answered by noting that it's not the first big project to take a long time (like, say Mozilla) to get done, though that challenges the ADD afflicted in the audience. Or it could be answered by noting that GHC, which is equivalent to Parrot, has a lot more time invested into it than Parrot does. (What, you thought the runtime engine sprang forth from the head of Zeus or something?) I could even point out that Parrot's been in fine shape to use for years. (I've been using it as such -- it's had perl 5's core functionality in it for two years or more, a lot less time than it took perl 5 to get that)

I could say that. Heck, if I were still in any position of authority with Parrot I would. I'm not, so instead... choose something appropriately dismissive that covers your ignorant cluelessness and go away.

Posted by: Dan at June 6, 2005 10:42 PM

From this conversation:
It looks to me like there is technical disagreement between DWIMery (Dan) and Orthogonal Computer Science Correctness (OCSC) (Leo). It also sounds like a very long running argument, and Dan is obviously very tired of fighting it.
Since I think Perl is very DWIM and TCL is very OCSC, I don't think I will like to program Parrot directly, given the current direction. Lets hope a nicely DWIM Perl6 can be created on top of the OCSC Parrot.

Posted by: Kris at June 7, 2005 02:28 PM

I am another of the original volunteers still around. :) (not for code though, merely infrastructure bits).

- ask

Posted by: Ask Bjørn Hansen at June 8, 2005 01:58 AM

Oh, and I too found the technical content (both on computer-ish subjects and cooking) very interesting. :-)

- ask

Posted by: Ask Bjørn Hansen at June 8, 2005 02:07 AM

Kris: Um... that's very much not how I'm seeing it.

Posted by: Piers Cawley at June 8, 2005 03:55 PM

Just wanted to add my voice to those wishing to express thanks for the blog - especially the 'What the heck...' posts, which finally pointed me towards the light on continuations (and a few other things) :)

As regards the state of the camel/parrot's health - I *really* hope that some kind of objective post mortem can be drafted so that those on the periphery of the community can understand what the heck's going on and try to learn from history rather than repeating it. More than anything else a key problem seems to have been a lack of open communication - and walking away in (relative) silence only seems likely to compound that. :\

Posted by: mjh at June 9, 2005 12:46 PM

Yeah, sorry to see you go. I like the "What the Heck" series, and it was interesting to hear someone else hack on the Mac.

Posted by: Jay at June 9, 2005 01:49 PM

thanks for the great "what the heck is..." posts

Posted by: timb at June 9, 2005 07:28 PM

thanks for the great "what the heck is..."

thanks for Dan!

Posted by: parrotsqlite at June 9, 2005 09:02 PM

Walking away in relative silence was the only way to handle things -- not walking away wasn't an option, and not being silent, well... there's a limit to how unprofessional I'm willing to be in public, and I wasn't sure I could manage within that limit.

The immediate reasons will be public in a bit, and a full-on review and lookback on my tenure (I hesitate to call it a post-mortem -- as I've been reminded, parrot's not dead :) is getting written up, albeit somewhat slowly. I'll make it public when it's done and been looked over by people more objective than I am about it all.

My inbox can't wait -- I'm sure it won't be lonely for long... :(

Posted by: Dan at June 10, 2005 11:42 AM

Another thanks. Parrot's a stunning piece of work, and the blog's had some neat stuff over the years. Hope you won't be abandoning technical blogging completely.

Posted by: Adrian Howard at June 10, 2005 03:09 PM

Thanks Dan for your gentlemanly behaviour and your contribution.

In the spirit of Elihu ( Elihus speech to Job ), please accept these words of encouragement from John Chatterton (quoted in 'Shadow Divers' Kurson, p 81)

1 If an undertaking was easy, someone else already would have done it

2 If you follow in another's footsteps, you miss the problems really worth solving

3 Excellence is born of preparation. dedication, focus, and tenacity; compromise on any of these and you become average

4 Every so often, life presents a great moment of decision, an intersection at which a man must decide to stop or go; a person lives with these decisions forever

5 Examine everyting; not all is as it seems or as people tell you

6 It is easiest to live with a decision if it is based on an earnest sense of right and wrong

7 The guy who gets killed is often the guy who got nervous. The guy who desn't care anymore, who has said, "I'm already dead - the fact that I live or die is irrelevant and the only thing that matters is the accounting I give of myself", is the most formidable force in the world.

(I'll leave the last one out).

Chatterton had Richie Kohler to encourage and believe in him. When one does world changing things like yours, such company is essential.

God bless you and your family.

Posted by: Stanley Hopcroft at June 14, 2005 09:24 PM

Wow. Squawks of the Parrot was pretty much my only view into the Perl 6 world, so this takes me by surprise.

I hope your enthusiasm for your craft hasn't taken too much of a hit. I know how depressing it can be to have something you're passionate about get squashed.

Posted by: Kimberley Burchett at June 15, 2005 08:57 PM

Nah, this hasn't squashed my enthusiasm for the craft, though that's in part because I don't have all that much enthusiasm for it as such. (I'm a writer and an architect. It's just that some days I write code and design software) Still, all is not lost -- I can always take my copy of the code as it stands and bend it to my will, something that appeals to the Mad Scientist in me. :)

Pity I didn't really key in on the fact that this was the primary source of perl 6 info a lot of people have. I feel like I missed this massive opportunity to spread bizarre disinformation. (Though I decided to discard my plan to try and convince people Larry wass a space alien from arcturus -- apparently the folks from procyon 3 take a dim view of those sorts of rumors about themselves...)

Posted by: Dan at June 16, 2005 09:21 AM

Dan said:
> I can always take my copy of the code as it stands and bend it to my will, something that appeals to the Mad Scientist in me.

Now you're talking.

Take some time to reflect and recover; then come back to parrot when you're refreshed - do it your way... on your time, at your pace, and in your style, with your sensibilities. Get help only when you need, and from who you choose/trust/respect to receive it.

Then parrot will fly.

Posted by: valis at June 17, 2005 06:04 PM

Dan, I was shocked to learn that you'd left Parrot. I've always been excited about VMs, and Parrot in particular, and I've learnt a lot from your writings (and presentations). I'd been waiting for Parrot to get finished (in the sense of being able to run Python or Ruby). Perhaps I'll have to wait forever...

It's incredibly saddening to see you go. If the Perl community keeps driving away people like you, does Parrot (or Perl6) have any hope for success? Or perhaps it will only be mediocre.

Posted by: Kartik Vaddadi at July 31, 2005 02:15 PM