some value of a few minutes

Well, after being down for some value of a few minutes, we are restored, and thanks to the generosity of the VCs it’s of course never ever going to happen ever again. Until next time. There is no such thing as 100% uptime, after all, even beyond the Land of Free. And shifting to shiny new servers and datacenters, while necessary for long-term stability, generally also entails short-term wonkiness.

Yes. I am the quintessence of glass-half-empty.

Also, I don’t think it’s fair to bitch at people for failing to make RSS backups (it inconveniences your subscribers and it’s far from an ideal format). Or to say that we have a backup plugin when we don’t. Or to shout at people when they say their blogs have gone forever. They only have your word for it about the backups. Why should they trust you? They might be refugees from diary-x.

So, on the one hand, I think that you really can’t expect any better from a free host and it’s up to you to take responsibility for your own data; and on the other, I have intense sympathy for people who perhaps aren’t savvy enough to work around the absence of a big BACKUP link, and then log on one day to find all their posts have vanished. Nobody knows in that first moment that it’s only temporary, or that they’re not the only one. The big BACKUP button might not be technically necessary, but psychologically it has real value. It cushions people against the fear that everything’s gone. It gives them back their content. It says you love them enough to set them free. It says you have enough faith in your product to believe they’ll stick around even if the door’s open for them to leave. Confidence inspires confidence. Saying ‘trust us’ at the precise moment the blogs have disappeared? Not so much.

Sometimes I think the reason we don’t have proper export is to give Movable Type something to feel superior about besides multiblogs. I can’t think of any other explanation.

14 Comments »

  1. Alan said

    In two years with my host, the only time I’ve had downtime is during a server move – and when it happened, my host gave me, and everyone else on the server, an Amazon gift certificate for being patient with them. I know WordPress.com can’t afford an approach like that (my host is boutique, it services less than twenty clients), but having the forum people snap at users who don’t know any better when something *does* go wrong doesn’t exactly shine well on the service.

  2. Ouch, I hear you. The problem is not in what happened nearly as much as it is in how it was handled through and through the whole experience.

  3. wank said

    To be fair, the only responses from an actual Automattic employee (podz, who really should have ‘key master’ ranking — AFAIK drmike is still working on a voluntary basis, though most people naturally assume he’s staff) were pretty calm and measured. Though from the special thankyou handed out to drmike we can safely assume his somewhat more… stringent approach had Matt’s full approval.

    Compared to the .org forums, the boards here remain relatively civil. Most of the time. So far.

  4. Amit said

    Sometimes I think the reason we don’t have proper export is to give Movable Type something to feel superior about besides multiblogs. I can’t think of any other explanation.

    Just to differ a bit here, when you say MovableType has a backup solution & compare it to WordPress.com, that’s not fair imho. The WordPress software does have more than one plugins for backing up database. So the ideal comparison should be WordPress.com & TypePad(about which I don’t know if it has a backup/export option).🙂

  5. wank said

    As Six Apart have always recognised, export is a basic feature which should be in the core, rather than delegated to third-party plugins. Even when they work, plugins aren’t always kept updated; developers get bored or busy, and they move on. In any case, plugins are entirely useless for wordpress.com users. We can’t install them so we can’t use them.

    Since you want me to compare Typepad instead, here goes. Typepad lets you export in the same format that MT does; it generates a text file which you can keep on your hard drive for backup purposes as well as exporting it into other tools (such as your own MT install, wordpress, textpattern, or expression engine). Unlike a database dump, you can browse your archives in a text editor or browser. You can import from Typepad into wordpress.com, but you cannot export from wordpress.com into Typepad without a lot of messing about with RSS options and converters. Oh, and losing all your comments.

    Nope, can’t see this disturbing 6A’s sense of superiority. Sorry about that.

  6. Alan said

    One of the biggest pains regarding the lack of export feature is you can’t actually move content between installs either. It’s all well and good locking someone into WordPress (or their comments/trackbacks or least – you know, the bit that makes blogging fun), but when you can’t even facilitate a move between WordPress installs, it becomes a pain in the arse like you wouldn’t believe.

  7. wank said

    Today we are promised an export feature in the next month or so.

    Given that back in January we were promised livejournal import within the week, my money is on ‘or so’.

  8. Amit said

    wank, I wasn’t supporting the “no-export” policy. Infact I’d very much like to be able to backup my wp.com blog so incase anything goes wrong, I don’t loose all that crap I’ve written over the months!!🙂

    what actually I was saying is that comparing wordpress.com and MovableType isn’t right since the former is a service while the latter is a blog software. So the ideal comparison would be WordPress.com and TypePad as both are similar services. I didn’t mean anything else by that. Be it that TypePad offers an export feature, I’m not arguing, kudos to them, I just thought that the comparison should be between similar things. WordPress & WordPress.com are different, similarly MovableType & TypePad are different as well.🙂

  9. wank said

    The impression I get is that Typepad isn’t quite as crippled in comparison to MT as wp.com is compared to wordpress; the more you pay, the closer it gets to having your own MT install. While wordpress and MT are targeting pretty much the same market, I don’t think the same can be said for wp.com and typepad. Typepad doesn’t offer a free service; it caters for serious bloggers who are willing to fork out so they don’t have to worry about tech stuff. WordPress.com, on the other hand, is a free host for geeks attracted by anything new and shiny, blogspot refugees, and casual bloggers who don’t want to pay for anything and aren’t bothered about their site looking the same as thousands of others. Not the same market at all.

    Either way, Typepad and MT are interchangeable in that they make it easy to export your blog; wordpress.com and wordpress are interchangeable in that they don’t.

  10. Amit said

    Typepad doesn’t offer a free service; it caters for serious bloggers who are willing to fork out so they don’t have to worry about tech stuff.

    really now, is there a holy writ that says that one’s gotta pay for having a blog to be considered a serious blogger? there are loads of serious bloggers(& popular as well) on Blogger.com and quite a number of serious bloggers on WordPress.com as well. just because TypePad users pay for the service doesn’t make them serious!! I can recall several bloggers hosted on their own domains(& paying for it) that I wouldn’t consider serious & I doubt someone else will. Same goes for people paying for hosting their websites & yet they are a total waste!!

    as for not worrying about tech stuff, exactly what tech stuff wordpress.com users gotta worry about? to customise their themes, they don’t have to dive into the template getting their hands dirty in xHTML or any template language, which both Blogger & TypePad users have to do!! You should just read about the gripes of TypePad users when the service goes down now & then owing to the resource hungry MovableType solution that it uses!!😉

    WordPress.com, on the other hand, is a free host for geeks attracted by anything new and shiny, blogspot refugees, and casual bloggers who don’t want to pay for anything and aren’t bothered about their site looking the same as thousands of others. Not the same market at all.

    I’ve read a lot of non-techie people agreeing that WordPress is quite easy to operate given its complex nature(not as easy as Blogger but then Blogger is as complex as a screwdriver). And as for the inability to have custom themes, well, to that all I can say is that this service is till in beta, improving everyday. Blogger & TypePad or any other service for that matter didn’t go bang out with all features on their launch day!!🙂

  11. Alan said

    I’m pretty sure TypePad offered customisable themes from day 1dot Being based off of Moveable Type, which houses all the layout info in the database, I don’t imagine it would be difficult for TypePad to offer theme editting at all.

    WordPress.com, on the other hand, looks like it’ll never have custom themes, as WordPress themes are written in PHP, and allowing people to freely access the PHP files that make up their theme, even if their intentions are harmless, could open up a massive security breach.

    It’s got nothing to do with WordPress.com being in beta. It’s got everything to do with security for the whole server.

  12. Matt said

    WordPress.com, on the other hand, looks like it’ll never have custom themes

    Do you know something I don’t? We’ve never said that. It’s non-trivial, but it’ll happen at some point. (It also may just be for paying users.)

  13. Alan said

    Obviously I don’t.

  14. wank said

    I’m pretty sure TypePad offered customisable themes from day 1dot

    Yes, because you can’t expect people to pay up if they can’t even edit their template. Theme customisation will be the killer feature which draws in paying customers and enables wordpress.com to turn a profit (the same way as livejournal found offering more userpics was the way to entice people into going paid).

    On the security side, I cannot see any reason for not letting paid users edit themes or upload their own, or even upload plugins. Once people have handed over their credit card details, they’re accountable in a way that those for whom you only have a disposable hotmail or gmail address just aren’t, and you’re in exactly the same position as any other host which gives its users code-executing privileges. You’re still taking a risk, but you’re banking on the fact that you know who these people are stopping them from doing anything stupid.

    Free users, in the meantime, could be placated by being allowed to edit their CSS, which is not quite as risky as giving them access to raw PHP. Livejournal scans all user stylesheets for dodgy-looking code, no reason why the same couldn’t be done here. It might make better business sense to reserve that functionality for paid users as well, though. Certainly it would be easier on the servers.

    Obviously they’ve learned from Typepad’s mistakes and wanted to get the server situation sorted out before taking anyone’s money (people get more stroppy about service failures if they happen to be paying for the service). And there may be some other features they want to get in place first (like export — paid users are not going to appreciate being locked-in any more than the rest of us). But I’d be surprised if by the end of the year people can’t pay extra for the ability to use their own theme.

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s