For those who don’t know, NaNoWriMo, short for National Novel Writing Month, is an initiative sponsored by the nonprofit Office of Letters and Light to get people to sit down and actually write the novel they’ve been saying they’d get to “someday”. I’ve done it almost every year for a while now, though I’ve never won. This year, however…
The point of Nano is to get to 50,000 words between Nov 1 and Nov 30. That’s it. The entire site is built around this concept, from pep talks to forums full of advice and resources to regions where you can connect with your fellow nanoers in coffee shop write-ins and so on and so forth. Furthermore, Nano occurs on fixed dates every year. The deadline for having the site up and running is November 1. The site needs some degree of functionality during October, but on November 1, it needs to be “gold”, or in a finished state.
This year, however, they’ve done a total redesign of the site backend thanks to the rapidly increasing number of users. I applaud this initiative; in past years, the site has been sluggish and unresponsive. Bravo. Unfortunately, they appear to have not budgeted extra time in which to do it.
On November 1, while the forums and basic wordcount updater were present so you could track your own word count, the API for external wordcount updaters, the widgets to display on the web (including the ever-popular “word war” widget that allows two scores to be displayed beside each other and the “region” widget that shows the stats for your entire region), the scoreboards showing how well various people and regions were doing, the wordcounts in the “gingerbread” on the side of the forum, and the “total words written” counter for all of nano were not ready. This meant the site had a simple word tracker with a buggy graph – something that can be easily reproduced in excel – and a forum with no special features – something that can easily be done in any of a number of forum hosting sites. There was nothing special or unique about the site itself except for the user base.
Meanwhile we’re being urged to buy nano-branded merchandise and donate. The donation drive was going full tilt, though some of the bonus features you get for donating and some of the store orders were going missing.
The Overall Words Written went live on November 10. But it was broken – it was showing 9 quintillion words written. I don’t have a record of when this was fixed, but I remember it taking a day or so.
Widgets went live on November 13. That’s 13 days behind schedule. More, really, because in previous years I’ve been able to put up a counter before the event began so it was up to date from minute one.
Staff Appreciation Day was November 15th. All over the forums, we were urged to thank our staff team for their hard work.
November 16th, the word count graph – the one thing that was ready on Day 1! – broke. On that day, no word counts input showed up on the graph. It was barely fixed in time for some people, depending on their time zone, to fix that day’s input; the rest of us have an inaccurate graph that shows 0 words written that day. There is no way to fix it, we are told, even from the backend. We’re stuck with a buggy graph. On that day, also, the widgets broke. Those were not fixed by midnight; the calendar graph, the one I use, is not fixed as I write this, five days later.
I donated in October. I received my first Doner Pep Talk on November 18.
November 19, a bug was reported in the API indicating that randomly, calls to the API would report “No novel exists”. This has not been fixed and likely was a problem before it was reported, as the reply was “this is a known bug”.
Several times we asked when these things would be available. Each time the answer was clear and standardized: they were not going to give any estimates or mention any dates because they could not guarantee they would meet them and did not want to be held to their own estimates.
By failing to create, test, and deploy core features of the product by the time the product was actually in use – in fact, by the time the product’s lifespan was half over! – OLL’s tech team have undermined confidence in their product and created a high degree of user dissatisfaction. When the site launched in October, several areas had “Coming Soon” graphics. These were never removed until the features launched; today, 21 days into the 30 day lifespan of the product, there are still “coming soon” bubbles (specifically, I’m staring at one where the scoreboards ought to be).
How did this happen? November 1st occurs on the same date every year – it’s not as though the deadline wasn’t known. The first bug was reported October 11, indicating that the site was live 20 days before the deadline. All the features listed were implemented the previous year and working perfectly, despite the site’s slowness. A lot has been bandied about changing the backend to Ruby; however, that had to have been more or less complete by Oct 11 if the site was functional enough to be launched. It took over a month to write the widgets? They’re implemented as image files located on the nano server. That’s all. They’re generated on the fly, probably every time the word count is updated. There was working code in whatever language the backend used to be; it’s just a matter of translating that into Ruby. But for whatever reason, 21 days after the project was due, it’s not completed.
I’ve been accused of complaining about “little things”. But in software development – particularly in web development – the “little things” are often all you have to distinguish yourself from the competition. Even when the competition is nonexistent, as with the case of Nano, that’s no excuse to be sloppy and unprofessional. Doubly so if you’re running on donations – when everyone can see the delays mounting, the lack of feedback, the lack of confidence in their own ability to meet deadlines, to then be asked for money and appreciation as a reward for the team’s “hard work” seems dishonest.
I’ve replaced the nonfunctional word calendar on the right with the simpler word bar because I have absolutely no idea when it will be working again. I’ve been using my own word tracking spreadsheet because I have no idea when the site’s word counter might break again. And I sincerely hope that next year’s project will be managed by someone who understands the needs of the community.