Crowdfunding the creation of the next version of LFGSS

Posted on
of 66
  • ah I was wondering why you were up! Sounding good :)

  • Sounds very exiting.

  • ah I was wondering why you were up! Sounding good :)

    Well, that's partly the timezones throwing my clock off. But if I have to be awake I might as well be coding.

  • ... with a view to creating an ugly first draft of the web app within 2 weeks of that date... meaning that we're currently under the belief that a very broken and ugly working version might be ready by the second week in March

    Does this mean LFGSS will be the Guinnea pig in March? With some live hacking and updating.

  • No, not at all.

    We'll run a basic version without the ability to import whilst we test a few things and refine it into a far more plausible and quality product.

    Then we'll run a private beta for a while for a few small communities who would like to test the software and are willing to start from fresh.

    Once we've done all of that, only then will we start writing the importers for things like vBulletin, Vanilla and phpBB.

    LFGSS doesn't get migrated over until it's a stable and quality piece of software, and the point of releasing early is to find flaws, test assumptions and rapidly improve the product into something the market (you guys and other forum admins) want before we launch it.

  • Dammit I was expecting to be a millionaire by now, where are my instant rewards?!

    (Seriously stoked about this, sounds like SF was awesome and has inspired you to really believe in getting the product made. Excited to see how the first draft goes)

  • Would TABS be able to to take part in the private beta?
    We are a pretty small community, as you know, who are looking form some forum software.

  • Yes.

    Once we are at the point where we are confident that we won't lose data and nothing too fundamental is likely to change, we'll open the private beta, and TABS can certainly be a part of it.

    It's private as in "not open to the world", but we're quite happy letting quite a few small communities try it out as the more that do, the more feedback we'll get.

    Without feedback this will all be for naught, as we won't be confident that we're making the right product.

  • Brilliant Velocio. Thanks

  • Typical TABS behaviour. Always trying to get into the primary position.

  • About existing communities that might want to move to microcosm.

    I run a BBS that used to be on vBulletin until I took umbrance at the behaviour of the company that acquired it. I then switched to SMF, which I loathe. I've got about ten years worth of posts there. If I were to switch to microcosm, what'd happen to them?

  • You'd keep them all.

    That simple.

    You could import them into a new forum using Microcosm, and then you're good to go. The existing users will need to re-confirm their email address post-import though... we haven't exactly worked this bit out yet, on some systems we might be able to make existing passwords work, but I don't know the SMF board stuff.

    I wasn't prioritising an SMF importer. But if you know what the export looks like we can take a look at it.

  • I'm not surprised - it's not amazingly popular software. The SMF team don't do nice tidy exporters with JSON or XML, so it'd be a case of mapping all the table data to microcosm's format. I'd actually be interested in giving that a go myself if you feel like letting me into the private beta. Might be an interesting proof-of-concept over how easy it is for external devs to work with.

  • It's not wholly surprising that they don't provide a tidy export format, LFGSS for example is almost 10GB of MySQL database. Can you imagine how inefficient that would be as a single JSON or XML file? And the likelihood for errors during import would be quite high.

    Once we're ticking over nicely, I want to make sure that we accommodate export in some tidy way to make sure you're not locked in. It would be great to help define a standard format for forum export (for the common data).

    The reason I like the thought of a common format is for data portability... what if every forum software could export in this 1 format? And what if every forum software could import this 1 format?

    Instead of every provider writing their own bespoke mapping thing, each provider would just write 1 export and 1 import.

    I think there's a lot to be gained by helping people move their forum and to try out different software. Not everything is a one-size-fits-all... so site admins should be able to try other things with reduced pain.

    And yup, happy to let you on the private beta when we get there.

    Not perfectly sure how we'll be doing import yet... whether we let you upload a database (but that requires us supporting every version of MySQL and Lord knows what flags and options), or whether we provide you with a script that you'd run on your server and that would call our API (most likely).

    Would be good to have your feedback too though.

    My only view on import is that it needs to be self-healing. That if something goes wrong it should be able to resume from the place at which it failed and once you've fixed whatever the issue is, that it picks up from that point.

  • Yeah, big DBs are an issue. I like the way Wordpress does its imports / exports - that works using XML but it breaks the imports up into chunks so if something fails halfway through, it's not a problem, it can just resume, going back to the self-healing thing. Uploading an actual DB has the potential to be a bit of a nightmare - what if your forum was running on MSSQL instead? And what about all the other weird data that people have, like VB arcade and so on. Realistically you'd want to go through the tables on the end client's server and just send the relevant data to microcosm instead of microcosm taking everything and then picking through it itself, so you'd end up with a bunch of connectors for different BBS packages, just like a normal forum, in a way people are used to.

    I do like the idea of a canonical, semantic format for BBSes, though. It would have been impossible five years ago, even in spite of microformats becoming more fashionable, but now that Open Graph has really taken off there's a real sense that all data can potentially be standardised and manipulated across the whole net in a really cool way.

  • I've imported a reasonably big SMF forum into a Drupal forum by hand before, unless you've got a load of extra plugins installed, the Schema isn't that complicated.

  • Hah... want to come and help us make the import/export stuff?

    BTW, one of the guys in the group is writing up notes on the SF trip. We (Matt and I) went as part of a group so that we would get an insight into how it all works, and Dan Hopwood is typing up his notes and putting them online, so if you want an overview of what we learned, start here:­6/lessons-from-across-the-pond-part-1-th­e-differences

  • I've imported a reasonably big SMF forum into a Drupal forum by hand before, unless you've got a load of extra plugins installed, the Schema isn't that complicated.

    I did Vanilla to vBulletin by hand for LFGSS.

    Well, I wrote the script for it. It all went pretty smoothly except for the private messages, as Vanilla stores these in the same table as the public messages with just some vague logic around the fields to present them differently.

    Moving 90% of the content (threads and posts) is generally easy. It's that last 10% that is the PITA. Though I am unlikely to every import stuff that 3rd party hacks introduced... meaning that we'll only really import 98% of data from a customised forum.

  • Hah... want to come and help us make the import/export stuff?

    I'll work out when my next Manchester-> London trip is, you were away for the last one.

  • I'd be interested in donating towards development for either an equity share, or for free use of the forum software depending on certain features (happy to chat through what these features are that I'd be interested in). Could possibly give some time from my developers too but that would depend on what systems you are using to develop in.

  • The forum software is designed to be free, no?

  • The forum software is designed to be free, no?

    Yes is the simple answer.

    Define free is the more complex answer.

    The primary plan we're working to is that the software and platform is free for everyone, and we take the affiliate revenue that may be generated. So it's essentially paid for by the users and their behaviour, rather than a fixed rental cost being charged.

    Though some feedback has suggested that a few entities (companies and local government) prefer the idea of paying based on usage to obtain SLAs and extra integration into existing sites, etc.

    Hard to know which is the more dominant view. So far free seems to be winning, and we haven't yet had a single site say that they're not prepared to give up their affiliate revenue to cover the inherent costs. Most sites realise that they never actually profit from affiliates, Microcosm only works on this basis because of the potential economies of scale.

  • And Bibble... I've emailed you.

  • Just checking you got my reply Velocio - more than happy to hear a "get the fuck out of here" reply if I'm barking up the wrong tree! It was late and I do tend to waffle on when I'm tired.....

  • Just saw this:

    Seen it? Seems similar but not straight competition(?) Some clever functionality in there, especially saved read states, auto follow, etc

  • Post a reply
    • Bold
    • Italics
    • Link
    • Image
    • List
    • Quote
    • code
    • Preview

Crowdfunding the creation of the next version of LFGSS

Posted by Avatar for Velocio @Velocio