Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie
[Suggestion] Auto-reply to PMs (out of office notification type thing)
Options
-
31-12-2011 7:30pm#1
Comments
-
-
-
-
-
It's because you're awesomeI've had numerous PMs from people thinking I'm a tLL mod. I never have been and I don't know why they think I am...
.
I'd say a good few people just see that someone is a mod on the left of their posts and assume that they are a mod across the whole site rather than just the listed forum(s).0 -
Advertisement
-
-
-
-
-
Moderators, Education Moderators, Technology & Internet Moderators, Regional South East Moderators Posts: 24,056 Mod ✭✭✭✭
Join Date:Posts: 21860
It is, which is why so much of Conor's 2011 was spent writing an API to the database to allow for the tentative first steps of that break. It's probably going to be 30-36 month project by the time it's all said and done (and we're over 6 months into that already). As it stands, new search and touch.boards as well as the "currently in development" iOS app are all bypassing vBulletin completely and are working off the API.
Is it going to be a custom board, or based on some other developed board? I just cant see how a few of the dev guys are going to be able to produce something as good, if not better, than vB so perhaps I am misunderstanding on what the aim of the project is?
0 -
Advertisement
-
Before I continue - I'll add a caveat that I'm not involved in the specific technical planning for this project, but am obviously involved in areas concerning front end functionallity and so on (which we're still in preliminary discussions on), so anything I say could currently be out of date or simply incorrect I'm sure if one of the tech team are passing they'll correct any errors in what I'm saying, but I don't want to get too bogged down in specifics on this just yet - it will be a big topic of discussion and the important bits will be explained properly to everyone.
It'll all be developed in house as far as I know. VBulletin is, and this is putting it nicely, so woefully and fundamentally inadequate for a site as big as ours that if the lads hadn't been already stripping out chunks of broken and dead code and functionallity, the site would probably have died some time around August. Search alone has had the site circling the drain for some time - it accounts for 12GB of the 16GB of RAM on each of the three database servers which is why old search will probably be permanently gone within the next 2 months.
It's not that VBulletin isn't good software or anything like that, it just doesn't scale to our levels of traffic and data and so we need to make the break.0 -
I just cant see how a few of the dev guys are going to be able to produce something as good, if not better, than vB so perhaps I am misunderstanding on what the aim of the project is?
Because we know our userbase, our server setup and the direction we want to take. Do you know the history of vBulletin? v1 was wrote by James Limm and John Percival, then came v2 which blew the mind of many an internet forum nerd (like me). v2 was programmed by 4 developers - Mike, Freddie, Scott and John. Guess how many Boards has? Modifying and hacking v2 is where I got started in PHP programming, and when I was younger I even sent an application to Jelsoft looking for a job. They advised me to go to college first and then come back to them. Even with all my vBulletin love going around, I still recognise that is it a one-size-fits-all solution and that it does have some significant drawbacks.
So why are we moving? Well that is a huge post that deserves some serious time as Dav said, but a small example on why is:
We don't need a special vBulletin "cron" call checking to see if background jobs need to be run - we have a system cron we can control for that.
We don't need X thousand lines of code for a calendar that nobody uses.
We don't need X thousand lines of code to determine if we're in a timezone that uses DST and if it's DST now.
We don't need X thousand lines of code for a language/phrasing system - English is fine.
We don't need the nightmare that is PHP/HTML hybrid code stored in a database without syntax highlighting/versioning called "templates"
We don't need a search that teeters the site on the brink of 503-hell
We do need increased flexibility to add features to the site that the users want.
We do need a better searching system, or at the very least a more efficient one that won't threaten to take the site out.
We do need a codebase that allows us to create more than the "Full browser on a PC" user experience (e.g. mobile-browser optimised site, iOS App, Android App)
We do need a codebase that will run fast and be efficient about it
We do need a codebase that will allow subtle background changes to happen to make the database more efficient
This is why Conor and Ross put a lot of work into new search, touch and the underlying API. Even though they've both moved on now they knew vBulletin was no longer suitable for Boards and the development decision was made to phase it out in favour of the API, our own codebase that we (as in Boards.ie Ltd) know because we spec'd it and wrote it with our userbase, needs and future plans in mind, knowing what we know about the last 10 years or so with vBulletin
In short: API = Freedom
0 -
-
Great post Danny.
Nope, API = A Potential* Independence_AVALANCHE_ wrote: »API = Application Programming Interface
*(or A Powerful Independence..)0 -
-
-
-
Moderators, Category Moderators, Entertainment Moderators, Sports Moderators Posts: 22,584 CMod ✭✭✭✭
Join Date:Posts: 21207
That's because Google can simultaneously throw a couple of thousand cheap cpu's at a search when it has to - there's no beating that._AVALANCHE_ wrote: »Google got in 0.2 seconds.
For that reason, I've given up using anything else other than Google sitesearch on the (comparatively minuscule) sites I manage.0 -
-
_AVALANCHE_ wrote: »Is that why is doesn't work all that well?
I was looking for a post the other night, couldn't remember who it was by, I could just remember abit of it, I went throught pagessssss (5ish at least:p) of Board search results, no sign of it.
Google got in 0.2 seconds.
Why not turn it off untill the API is ready?
The API is already in production, otherwise new search and the Touch site wouldn't be online and the iOS app wouldn't be anything except a fancy interface. It is obviously nowhere near feature complete, and who knows if it ever will be - Boards is constantly evolving
Search does have some minor issues mostly related to uptime but let me state this here and now: It won't ever be able to compete with Google. Why? Because we simply can't. We're not a search giant, we don't have hundreds of developers working on search algorithms, natural language processing or server infrastructure to give you better, faster, more accurate search results. What we DO have though is the ability to search all posts on Boards.ie paying attention to your permissions. Google has as much access to boards as a guest who isn't logged in, you'll never find a private forum post on there. You can't order Google results for Boards by "thanks", or by author, by thread started, by last reply or by forum. However Boards search will do all of that.
If you want a general keyword based search over public forums only then fire away and use Google if you wish but don't underestimate the power of Boards search - it was done before I arrived here but considering what was involved and the hardware under it some serious reviews took place and the team had the stats/metrics of our old search to justify the need for a new one instead of relying on Google0 -
Advertisement
Advertisement