February 2001. Opera 5.02 was the current Opera release, 5.10 was being worked on. The ECMAScript engine was called "linear A" (to be followed by linear B, Futhark and Carakan). And Jóna Björk, head of Opera's customer service department hired a new employee with the hard-to-spell name Hallvord.
That makes ten years. A long time in tech, a milestone to mark – yet it doesn't seem that long ago really..
I thought I'd post some stats, for a dry and geeky anniversary celebration..
- Support requests replied to: over 10 000 (the old support system is no longer used, so I can't say exactly)
(I should go have a look at the seven "inactive" ones shortly..)
- Bugs I'm CCed on, watching or QA Contact for: 13640. Here's a chart of the ones I'm CCed on:
New, examined and (if I have time) inactive are basically on the QA team's table for analysis. There is some work to do.. 🙂
- Sick days: none. According to the system tracking such things, I haven't been ill for ten years. Wow! 😎 It's not entirely true, but because I choose to work part-time, Opera gives me such flexible hours, and I work from home about 1/3 of my time anyway, I've always been able to catch up with the hours I should work.
What has changed in 10 years? The web has of course changed enormously. But, as they say, the more things change the more they stay the same. Opera retains its peculiar blend of stunning excellence and stunning shortcomings. It has more users than ever but in certain important markets still isn't counted as a serious contender.
- Every released bug, especially bad regressions in a new, shiny something.00 release.
- Switching to Maconomy's ERP system. The older homegrown time reporting system may have had its limitations, but at least it was simple and fast. What was even worse than having to use the new system: shortly after we started using their ERP, the Maconomy corporation was advertising all over Oslo Airport that Opera was their customer. I hated every single ad :(.
- Browser sniffing. Say no more.
- Implementation of browser.js. It was spec'ed and implemented pretty quickly, mainly to fix the dynamic menu on sony.com. They used at the time a menu script called Ultimate Dropdown Menu, and one of its features was to fall back to a plain UL/LI based HTML version if the script didn't work. Unfortunately, UDM 4.3 contained browser sniffing that only worked with Opera 7 and failed with Opera 8. The menu appeared as an unstyled list, pushing all the actual site content several screens down – not a pretty sight – and it was next to impossible to get through to anyone at Sony who could do the simple upgrade to UDM 4.4. This kicked the discussed browser.js to the top of the priority pile, and I guess neither web developer James Edwards aka brothercake nor Sony's web master knows what influence they had on Opera 8.01's feature set.
- Taking web compatibility more and more seriously. With all the sniffing and broken code on the web it's both natural and tempting to point fingers when users have problems, but it's way more productive to realise that there is also a considerable pile of problems that are caused by bugs or choices we could change. Today we have a more systematic approach to bugs that break major sites than ever before. It still has a bit of a whack-a-mole feeling but we also know that the dead moles are piling up. Hey, if I wasn't an optimist by nature, I probably wouldn't still be doing this job after ten years.. 😉
Finally: thanks for using Opera some or all of those years, thereby making all that hard work possible :D. And if any readers aren't using it yet, go get it.