IPC Report

Posted November 21st, 2007 by

At the beginning of November I attended the International PHP Conference in Frankfurt, Germany. This was my first attendance at any of the larger conferences and I thought I’d share my experiences as a conference newbie. The most noticeable thing was that all kinds of minorities were wildly under-represented. There were very few people over 35, female or with non-caucasian appearances. On the women front I counted about ten, excluding organisers, and I’d say most of these were attending the AJAX strand of the conference – on the PHP front there were only two female speakers.

The hotel itself was a really nice venue, and well-suited to an event of this kind. Between talks there were plenty of drinks and coffee laid out for everyone, and the snacks which were provided in some of the breaks were really tasty. Their meeting rooms were well-equipped with projectors and so on, however the wireless network didn’t survive the strain of so many geeks in a small area and was intermittent at best. The “expo” was about 5 stands clogging up the entrance hall, with the main sponsors each having a stand. EZ Systems, SugarCRM, Code Gear, Mayflower and some german-language publishers were there, and there was a Zend Lounge in the corner consisting of three sofas.

The night before the main conference, there was an event called “Code Camp @ Night” where two internals developers (Marcus Boerger and Johannes Schluter) delivered a workshop on extending PHP. The session also featured pizza and beer which I saw as a major selling point. The website listed the session and said that it was imperative to register for it, however there was no means to do so on the site and my repeated emails to the organisers met with no response. Since we weren’t registered on the conference until the following day we didn’t think we’d be able to attend but we dropped in to the hotel a day early and were told we’d be welcome at this event, which was great news! The event itself was really interesting, if a bit over my head.

Pizza aside, the food throughout the conference was extremely good. The days were very long, with two meals being served on the Tuesday and a main meal on Wednesday. It was a hot and cold buffet setup with large dining spaces – I think everyone managed to find something they were happy to eat, with fruit, salads and dessert being available as well as a selection of hot food.

One thing I found strange throughout the conference was that almost all the attendees and speakers had German as a first language. There were sessions in both English and German but even the English sessions often started with “does anyone NOT speak German?”, with the speaker hoping to be able to communicate in the majority language of the assembled group . Since I only speak English I felt quite awkward about the whole thing and having sessions in both languages didn’t really work that well for me as there were quite a few that I couldn’t attend. The keynotes were all in English and perhaps it would have been more straightforward to have everything in one language or another – in the end I had to go through and highlight which sessions I _could_ attend and then choose from those.

The keynotes were of varying degrees of interest and quality. First up was Zak Greant with his “Age of Literate Machines”. This was a fantastic and inspiring talk, Zak was eloquent and interesting and his slides were excellent. His point, that only open systemns should be used for government of society in order that everyone can understand the mechanism that controls their lives, was noble and thought-provoking. The following day Derick Rethans’ talk “Personal Home Page Tools Have Grown Up” showed a clear (and amusing) history of how his experience has grown alongside the language and how much he and it have developed along the way. The final talk was much less interesting, Bill Scott on “Design Patterns and Principles”; this didn’t involve design patterns but instead revolved entirely around inaccessible, client-side technologies. I’m sure the point of this talk was to cater to the AJAX conference that was happening in tandem with the PHP one, but it didn’t do anything for me and I hope it wasn’t representative of the content in that strand.

The sessions themselves were generally very good, with four to choose from at any given time. On the first day I think I picked the mainstream topics since quite a few of the talks I attended were standing room only. I can’t speak in general about the sessions as of course I only attended a small sample of the choice of sessions that were available, but what I did see was very interesting, put across well, and were well-prepared which is always a bonus!

There were long breaks between the sessions which was excellent and allowed for plenty of chattering, coffee drinking and table football (I lost count of the number of times I was asked the English word for it!) and enabled everyone to mingle. On the Tuesday evening, the middle night of the main conference, there was a Casino Night, with Poker, BlackJack and Pontoon tables running with a free bar. The games looked like fun but basically I couldn’t stop meeting new people for long enough to try them out! Certainly it was a good ice-breaker and people who didn’t know anyone could sit at the games tables and get talking with others as the evening progressed. To add a finishing touch to the conference I was also able to tag along to dinner in a local eatery with a bunch of speakers and others who were staying over the night following the conference – talk about being in with the cool kids!

All in all I am really glad I made the effort to go, I met quite a few people in person that I knew of by reputation or had met online. Having never attended a conference it was interesting to see how things work, see different styles of speaking and presenting, and generally get a feel for these things. The party atmosphere I heard so much about from ZendCon and the php| conferences was definitely missing from this one, however I did have a good time regardless. Would I go again? I don’t think I would, I paid for the conference and the trip myself since my employers didn’t see it as worthwhile, and for two days I must say it really did seem like a lot of money. I won’t say it was a bad conference, and I don’t have a lot to compare it to – certainly there is nothing else close to it in Europe. Perhaps if I wish for it enough the rumours of a European ZendCon will come true one day …

Posted in Articles, News, php | 5 Comments »

Interview with Zoe Slattery

Posted November 13th, 2007 by

Zoe’s talk at the conference was entitled “TEST || die”; she talked about how she got into testing PHP itself and her hopes for the future. There is little activity on the PHP QA mailing list and a great need for more tests to be written so that changes to fix known bugs or add new features will not break existing functionality. She is disappointed that more people aren’t already involved in writing tests as it’s a really good entry route into being more involved with the core of the language and the entry level for it is much lower than submitting patches for PHP itself. I caught up with her the next day to ask how she ended up here and championing this cause!

How did you get started with PHP?
I was a manager at IBM for some time and the nature of my job meant I had little time for coding. I picked up PHP because it is easy and quick to learn when I wanted to write a tool.

Why did you start writing PHP tests? What made you decide to help PHP in that way?
One of my colleagues (Graeme Johnson) went to the PHP International Conference last year and met Derick and Marcus there. Graeme came back with the message “PHP needs more tests”. I was just moving out of mangement and had 9 months to fill before I went to University – I needed a project that was going to be useful to someone, self-educational and that IBM would be happy for me to do. Writing PHP tests met all those criteria. I’ve always been a better tester than anything else – I think it’s a combination being able to see the way users will do something and be able to read code enough to understand what the developer meant. I suppose that other factor that influenced my choice to do this was that I’m a fanatical supporter of open source development …. but that’s another subject :-)

You’re taking a year out, what are you up to?
I’m going back to university, taking an MSc in Software Engineering at Southampton University. The course will cover Java but maybe I’ll get to do PHP as part of my course as well. It’ll be good to bring my technical skills (which got rusty when I was a manger) up to date with subjects like OOP.

Have you made contact with any groups in the PHP world?
I do keep up with the mailing lists – internals, QA and the CVS list. I do also sometimes use the #php-pecl channel on IRC. In the real world I go to the London PHP User Group meetings as well, they’re really interesting. There is a broad range of people there and you find out which companies are using PHP. People often show off their various applications there too, I’ve seen some really good ones.

Do you have a particular support network?
When I started out writing the tests for PHP I did feel really alone. Once I started submitting tests though, I started getting feedback from the internals guys like Marcus Boerger and Tony Dovgal. It’s really scary at first, working with people that are very expert developers working on complex software … but they really appreciate the input and responded well.

How did you come to be speaking at this conference?
When I saw the call for papers I submitted the proposal and they offered me a quick talk slot (15 minutes in the evening). There was nothing special, I just thought the subject was important and went for it.

Have you got much speaking experience?
The last time I presented at a conference was 30 years ago! I was really nervous but actually I quite enjoyed it once I got there and I have had lots of practice at presenting inside IBM and to IBM’s customers – but somehow it’s not the same.

Can you tell us a bit about your talk for those that weren’t there?
The point was to raise the profile of testing PHP. We really don’t have enough test coverage for PHP, but people don’t see it as an area they can contribute to. It’s seen as being difficult but its not! Maybe because I’ve managed a lot of software development I have more of a “bigger picture” awareness, it’s really clear how important this is to the future of the PHP project.

You’ve had a long technical career AND children, how has that worked out for you?
I have always been determined to have both, even though having dependent children of any age makes it hard to concentrate on anything else. I have twins who are now at university and I’ve worked for IBM since they were 10 months old; it helps that my husband has always been around and done his share. When I went to work for IBM I made arrangements for working hours that fitted in with my family and I stuck to them. I don’t know what my colleagues thought of it but it worked well and IBM are a really good employer. They have fantastic maternity arrangements and also make a big effort to bring in women from all sorts of backgrounds and to raise their awareness of the possibility of a technical career. I think being a woman can be an advantage – people will always explain things to you, and we’re better at asking for help.

You’re trying to get people writing tests for PHP – where do we start?
Go to qa.php.net and there are instructions on what is needed for a test for PHP. To find out what areas need tests writing, visit gcov.php.net for the current test coverage of the various libraries. We are considering making this simpler by having a list of testing priorities that people can pick from. At the moment you have to submit tests to the QA mailing list main PHP CVS repository if you have access, but there has been discussion around making an intermediate “safe” area to submit your tests for review by other testers before they go into the main tree. If your test is then used, we’d let you know and make sure you were credited for your contribution.

You’re trying to single-handledly get PHP tested and overhaul the process for submitting tests. Do you know what you are getting into?
Yes! But I really do feel that this is a really, really important factor of the project, and trying to help is important. Actually – it’s not single handed – my colleague Raghubansh Kumar has submitted over a 1000 test cases, Rob Nicholson has added some too. Tony Dovgal, Marcus Boerger and Derick Rethans have all reviewed the tests and made helpful comments, Nuno Lopes fixed run-tests.php pretty nicely too. So, I’ve had lots of help.

Posted in News, php | 4 Comments »