Ok, that title is alarmist and a straw man. Just chock it up to my catching up on slashdot articles. I guess its time for a rant of my own.
In the past week I have been told a three times that I am the ‘voice of PyCon’, and others have viewed me as speaking from some position of authority (their exact word). This frightens me, as I really see myself as just another python user and like any other PyCon attendee; I just happen to have volunteered my time, like many, many, others. I still use perl regularly, program primarily in C++ on Windows, and lack minions of any kind (though I am taking applications). So before we get into this mess any further, I would like to make it clear that I am not speaking on the behalf of the ‘PyCon’ (as if it had somehow become a thing incarnate), the other organizers, or the PSF. I am speaking for myself as a sponsor, a volunteer, and most importantly as an attendee. These opinions are my own.
Unless you live without an RSS reader (the ‘under a rock’ of the web 2.0 world), you are aware of Bruce Eckles disappointment with PyCon2008. If you have not yet done so, please do read the entire thread, as there is a wealth of useful information in there. I did notice some minor personal attacks against Bruce, which are completely uncalled for. The lightning talks on Friday and Saturday were a mistake. Mistakes happen, and we must learn from them. With the alarming growth of both python and the conference (<400 in ‘06, <600 in ‘07, and >1040 in ‘08), there are bound to be growing pains. Preferential treatment given to sponsors was a failed experiment. For a full explanation, one should read Van Lindbergs’ post on the matter. Problem solved. I do not agree with the other assessments Bruce made. (DISCLAIMER: I was the sponsor contact for my employer Nuance Communications. I did give a sponsor lightning talk. I believe it was well received.)
Bruce expressed a disdain for the sponsor diamond keynotes. Personally I thought they were fantastic. They might not have been very technical, but they were fun, enjoyable, and very ‘pythonic’. These companies are part of the community too. Ignoring them, their presence, and their influence on python does not diminish their impact in any way. We need to welcome them as the active participants in the community they are, and call them to task if they ever attempt to put pressure on or subvert the community. I know of no organization, sponsor or otherwise, that has done this. Two 15min talks on why these companies are sponsoring and supporting python and the community is something I personally want to hear. If these talks were bad or showed some unpythonic bent, I would see this as a negative, but that was just not the case. I can’t wait for the video to be released to match the comments to the reality.
NOTE: so far we are talking about <3 hours out of 48 hours of core conference time (much more if you consider parallel talks as serial time, which is defiantly the case for the AV crew. I do not have the end number, but we have well over 100 hours of usable video to process.) Three very crucial hours of PyCon were lost. Three whole hours. (Well if you just talk about the ’sponsor issue’.)
Bruce also expressed an opinion that the quality of the talks this year was well below the average for PyCon. Others have also expressed this opinion, and some have expressed aggravation with the program committee. I have learned not to take this personally. Aahz wrote a self titled rant placing the fault for this on the attendees and the community. (He later clarifies the distinction between ‘fault’ and ‘blame’ which is critical.) This was defiantly an impassioned, over the top, melodramatic rant. Usually it is ME making these rants and Aahz calling me out for it. Placing the fault on the attendees the way he did was uncalled for, and he knows it. The truth is there are thing the program committee can do to to make the process better, and many of these have been discussed over the past two days; but that is a different post. There is a core truth to what Aahz says though. We put out calls on c.p.l., the organizers list, and two blogs for help in the program committee, as well as on the pc-list. This resulted in 8 people dropping out of the list, and two new people stepping forward. We targeted people in the community but they were all ‘too busy’. I can respect that, and do not hold it against those people. But we do need the community to step up. At some level, if PyCon no longer represents what the attendees want, then they need to be the ones to correct it. No one will correct it for them; constructive feedback withstanding. Griping is fine, helping with the solutions is better and much more productive.
There were 2 other misconceptions (lies) on the thread about the program selection that I feel need addressing as well:
1. Reviewers did not consider author names/credibility in the selection process.
Bullshit. Of course we did. Initialy we did not see the author names, but as part of the process you had better believe we considered it! One person felt they should get their talk accepted on name alone in ‘06 as they did not have time to answer questions by the reviewer and refused to update their proposal as they were too ‘busy’. Good, your talk should not be accepted. If you cant take the time to put in the effort during the review process and decide to completely ignore the program committee, then I really don’t care who you are, you have shown that you do not have the time or commitment required to be a part of PyCon. If you want your time to be respected, then at least have the decency to respect the time of others. As a result of this they did not submit a proposal in ‘07. What a great way to take part in your community conference.
2. There were not ‘beginner, intermediate, advanced’ distinctions on the talks.
This text was not printed on the physical timetable handout (1 huge page), but it was included in the 68 page full color program guide along with the talk synopsizes and on the website all over the place. I guess next time someone will get up and read the program guide as a keynote. (This might be a great Idea as there is a full description of the different parts of the conference in there including what and how open space works which is something else people complained was missing)
If there is a silver lining in all of this, it is that people are stepping forward to help on the program committee and help with next years PyCon; I will believe it when they show up on the lists. Many are also saying that we should have more unconferences and python code mashes. Many people have voiced this, and all I have to say is ‘about fucking time!!!!’ Just remember, no one is going to do it for you just because you post in a mailing list that it should be done. Get out of your mail client and go do it!
One final note: There were more than 270 people sprinting today on 22 open source python projects, more than attended the first PyCon, and almost 4 times the number of people as last year (86). PyCon’s failure as a community conference has been greatly exaggerated.
UPDATE: fixing a spam attack broke the title. Fixed now.
We seem to be discussing this all over the place!
“But we do need the community to step up. At some level, if PyCon no longer represents what the attendees want, then they need to be the ones to correct it.”
I think this says it all. There’s no point in people claiming that the community wasn’t represented if they, as the community, didn’t seek to represent themselves. But then I think people need to be continually reminded and/or introduced to the idea that they have a stake in the event: it’s not “pay to view”; it’s “pay to play”.
If you travel all the way to the conference, it’s best to make the most of it, and it’s worth noting that Bruce, despite his rant, did seem to be participating heavily in his own way in other aspects of the conference, from what I’ve heard at least.
The only valid point in Bruce’s rant is: the quality of the talks was really poor this year.
I have been in talks on all three days and compared to the last two Pycon conferences in Dallas there were almost no talk where I would have said “wow, that’s cool”. Lots of boring stuff..I am a bit disappointed and have to re-think making the travel from Germany again next year.
Well said.
Your mention of Slashd*t at the top of your article got me all excited that maybe PyCon had finally been mentioned on that site after all these years (even if in a negative light, that’s *something*) but sadly no. It looks like we’ll never get onto the “news for nerds” site
Very nice, Doug
I’ll add a couple of points:
Yes, the lightning talks on Saturday were lame and boring. I don’t particularly care who’s “fault” that was, so please feel free to blame me — I’m wearing my flameproof pants. Point is it was a mistake and will be fixed next year. Did anyone attend Sunday’s talks? I thought they were *much* better.
Yes, I too thought the quality of the talks were subpar. Again, I blame myself and will be on the program committee for 2009 if the organizers will have me.
Finally.
If you didn’t like something about how this year worked, volunteer to help out next year or shut the fuck up. Thank you.
[Doug, feel free to edit for language if you so choose.]
@Andreas:
I actually believe that Bruces’ 3 main points are all valid (even if I don’t agree with all of them). The lightning talks were a mistake. I loved the sponsor keynotes, but there is a risk in doing that. Just because I loved all the talks I saw, does not mean the talks as a whole represent the community, as I alone am not ‘the community’.
@Jacob:
Please do not blame yourself. Every year you help with the organization of the conference and are a fantastic lightning talk coordinator and sprint leader. Van is already trying to take on the blame for every little thing that could be conceived as going wrong (which is just silly). We need more people, not the same ones spreading themselves thinner.
Normally I do edit language, but I too used some choice words, and see it as a form of public therapy. Thus they remain intact.
@Andres:
I wanted to respond directly to the part of your comment about being stuck in boring talks.
I am very sorry that this happened. I hope that the video’s of the other talks you missed will give you back some faith in PyCon. Even if you do not come back next year, please consider helping the program committee to ensure that others do not have the same impression you had this year, next year.
What might help a lot would be to release the survey results for liked/disliked talks to the actual speakers who attended.
As one such speaker, I’d love to hear if people had feedback on the talk. I spent months preparing just the right talk and I’d love to know how I can improve to make sure I get a chance to talk at future Pycons where I can add value.
I have to add that I really did enjoy this Pycon.
I think an electronically controlled open space board would have solved a lot more problems since people uncomfortable with the less interesting talks would be able to quickly and easily review their open space options while sitting inside a session.
Another cool idea is to have a open space “lightning intros” going on during lunch or other down times to help give people a feel for the literally “underground” sessions that were going on.
Jacob Kaplan-Moss wrote “If you didn’t like something about how this year worked, volunteer to help out next year or shut the fuck up.” That same sentiment has been expressed in other forums in response to criticisms of PyCon.
From my own experience I understand the frustration that comes from volunteering for a job, doing a lot of work for other people, and then having to listen to complaints and criticisms from those who didn’t volunteer.
That’s how volunteering works.
If you can’t take the criticism, don’t f**king volunteer.
I appreciate the PyCon volunteers weren’t paid and put in a lot of their own time, for their own reasons, but that doesn’t mean the people who paid to attend and took time off from whatever else they have to do should just shut up or join the group of volunteers next year.
The registration web site couldn’t take credit cards for days on end, and crashed repeatedly with uncaught exceptions (clue: use working PHP payment processing code next time). Trying to book and confirm a hotel room seemed to be handled by one person on a cell phone at Starbucks. There was no phone number to call about the credit card problems, just responses from personal gmail.com email addresses. There’s no reason “run by volunteers” has to mean “amateurish.”
I don’t think we all have to volunteer to earn the right to criticize the conference. Simply paying for it gives attendees the right to complain. As Dr. Johnson put it, “You may abuse a tragedy, though you cannot write one. You may scold a carpenter who has made you a bad table, though you cannot make a table. It is not your trade to make tables.”
Imagine the size of the Python user base if every criticism and complaint about the language was met by “f**k off or do it yourself” response. Get a grip, take the criticisms in stride, and if the job is just too thankless for you, don’t do it, but curse and insult the people who had every right to expect a better PyCon experience.
@Greg:
Thanks for all the straw men!
0. yes Jacobs’ comment is a bit over the top, but you get the point intended. Don’t be so literal. Get a clue.
1. No you do not need to volunteer to criticise the conference!
We are accepting the criticism and making changes in response. as I have said MANY MANY times now. ‘Griping is fine, helping with the solutions is better, volunteering is best.’
Helping with solutions can just be submitting criticisms with ways to solve the problem, which many people have! It’s fantastic!
2. As for the registration system. Well how about I give your number out. Are you up for that? No one stepped forward to take on that level of responsibility. If no one volunteers, it doesn’t get done. period. I am very happy you have stepped forward to take this on.
3. No one is not taking the criticism. Where do you see that? Every one agrees that things did not work. Personally I Loved all the talks I saw (sans the lightning issue). But that is my opinion. I think the PC did a fantastic job. If others don’t feel that way, then they need to join the PC to fix it, as I personally see nothing wrong. That is the point. If you don’t like it, and the people who did the work do, then the only way to fix it is to fix it yourself.
I will not dignify the “Amateurish” comment with a full response. If that is how you feel this conference was run, then there is nothing more I can say on the matter.
While I do wish that the program committee had the author names open for the whole process, I otherwise agree with what you have said, Doug. People who have not attended multiple PyCons might not have the proper reference to realize how much we truly have grown since 2007. And it’s unfortunate that some people are going beyond just saying “I think this could be changed” to “the PyCon organizers failed” or some other form of blame.
There was really no way to predict how much Pycon would grow this year. Original estimates at the end of last year’s Pycon, IIRC, were for an attendance of about 700.
The same held true for vendor participation. It seems like this year everything, including Python’s user base and commercial support growth, just exploded.
Although I did little to help, I did follow the Organizers’ mailing list throughout the process. Not only did these folks work their a**es off, they were quite competent and on top of things as well. Major holidays were spent hashing out organizational items. People were up late at night on weekends working to make things happen.
I think Pycon just has more momentum right now than the communtiy was ready for. Next year the registration and sponsorship blitz won’t be (as big) a surprise.
My 2 cents.
Carl T.
@Greg:
Note on the payment system being down and crashing. That was NOT our code or our site. That was our payment processor ‘PayPal’ for which we are spending thousands on that system. Using the PHP cart’s out there which integrate with paypal WOULD ALSO HAVE CRASHED!!! As it was not our system at fault but paypal’s. Thousands of sites were effected by those outages.
Don’t believe me? Well here is the proof!
And the problems have continued intermittently afterwards as can be seen by the status feed.
If anyone was ‘amateurish’ it was PayPal. Go figure. Our system did not loose even ONE registration even with all the problems. It never, ever crashed even once. THAT is a solid system.
So basically to everyone who thinks our registration system sucked, I can say with a clear conscience that yes it did, it was not our fault. I am better developer than that thank you very much.
Doug,
Volunteering for anything can be a thankless job, as I know from my own experience. Lashing out at those who didn’t volunteer with profanity, sarcasm, and snide arguments only makes the situation worse. Any inclination I might have had to help out next time is gone now that I see the kind of thin-skinned attitudes some of the PyCon team has in response to criticisms.
That said, clearly everyone who volunteered deserves thanks regardless of the problems.
As for the payment system, the problem is the PyCon site used PayPal’s post-back system, which turns control of the transaction over to PayPal’s servers, thus leaving customers exposed to a system out of your control. The right way to authorize credit cards is to send the transaction from your server to the PayPal/PayFlow Pro servers, behind the scenes, so the user’s browser doesn’t ever leave your site. You can use curl, for example. That way you can trap errors and problems and show your customers something more meaningful than a Java stack trace. I have quite a few clients using PayPal as their payment processor, and some of them experienced problems during the PayPal outage you pointed to, but none of them dumped customers onto PayPal’s site to see raw error messages. Your e-commerce system should not leave your customers exposed to unreachable or broken payment gateways — that is not state of the art and is, in my opinion, amateurish.
It’s a shame PayPal still supports the postback setup; Verisign was pushing customers away from that before PayPal acquired PayFlow Pro from them.
And if PyCon really spent “thousands” to process credit cards through PayPal, I’m pretty sure you got ripped off.
Request:
a moratorium on use of profanity and the word amateurish.
Where’s the love?
Greg,
The volunteers needed this information months ago. If you’re like me and don’t have the time to contribute, your lurking on the Organizers list sure would have helped out.
No matter how you feel about the organizers now, give it some consideration. That whole PayPal discussion, IIRC, lasted at least a week. Your intelligence (both in the brainy and military sense) could have saved people a lot of trouble.
Thanks for taking this input into consideration.
Again, my 2 cents.
@Greg:
I agree that lashing out with profanity and snide comments is not the best reaction, but sometimes it sure feels good to respond to snide comments about using PHP for a Python conference with them. For that I do apologize.
Again, you are correct that Jacob’s comments were over the top, but if you read around, you will see that the organizers are responding to the feedback in a constructive way. As I mention in my post Aahz’s similar comments that attendees are the ones at ‘fault’ if the conference was not good, is wrong.
As for the PayPal system, you are correct that curl does support doing that, but the newly implemented ’security’ features of paypal prevent one from doing this as it will ‘lock down’ the account when too many requests come in from the same IP. They are moving everyone over to a SOAP interface. I did not have time to write a complete SOAP shopping cart, nor did we have the money (at the time) to pay for one of the approved carts which start at $4K. Those are an option for next year. There are some very nice Python based shopping carts which we are evaluating.
Yes we were ripped off. We will be getting off of the system. We did not discover any problems until they started, at which time it was way too late to switch.
I find it a little closed minded that you decided not to come to PyCon because paypal crashed on you, and because you failed to read the instructions on the hotel registration page, but that was your call. I did take many of your comments, especially your tone, both here and on the registration system, a little rude. It does appear that you have no intention in being part of of the python community at PyCon. That is also fine. As I said previously I do hope you enjoy the other conferences you go to. That was not a snide comment, but an honest sentiment.
You know, I had an awesome time. This is tough-earned praise, because my work has been extremely busy and I’ve been having to make some very tough choices about where I spend my time, including not attending or cutting short things which I love.
I gave PyCon 9 days that I didn’t really have.
So, when I say that I’m looking forward to PyCon next year, I consider that to be pretty telling.
I think PyCon in general (and specifically the volunteers that made it happen) did an exceptional job of handling 50% more people than planned for. PyCon this year was huge, and that alone made it very hard to handle.
Sure, there were weak spots. But, I made the most of them.
So, lunch wasn’t as good as other years, but having it available made it so that we didn’t have to fart around going out and so I had more opportunities to interact with other people.
So some of the lightning talks were total corporate marketing… I turned to my right and chatted with Richard Freaking Jones.
What did I think about the talks? I had a whole slew I wanted to go to, and in particular 4 or 5 I really wanted to see. I saw two of these, one met my expectations and one blew them away.
But in general, I was too busy at the community conference interacting with the community to go to the talks. My general opinion is that I can learn what they’re talking about at any time, I don’t need to travel to Chicago to do that…
I come to the community conference to interact with the community. And PyCon had that in spades.
I made the best of something that had, admittedly, some problems. Would I have rather paid 10 times the money for one of these massively polished conferences? Absolutely not. I really don’t like the overproduced conferences. Amateurish? Sure, that’s because we’re all amateurs at putting on conferences. Which I think gives PyCon a lot of it’s appeal.
PyCon isn’t meant to be the Java 1 or Comdex of Python. It’s a very specific thing and I think it’s good at what it is.
The best talk I went to? A group of around a dozen Twisted folks and others sitting in a circle talking about licensing at midnight in the Deli.
PyCon? It’s a synonym for awesome.
Sean