Albion stats website - designer wanted

[The History Man]

From time to time there are proposals on NSC by various people for an Albion statistical website. Brighton & Hove Albion Collectors' and Historians' Society, which has all the match statistics - including line-ups - required, has also had ambitions for several years to provide such a website, and has owned the domain name www.AlbionHistory.org.uk for some time.

So it's about time I pulled my finger out and tried to get such a site up and running - but I don't have the time to attempt the necessary programming myself, I'm afraid.

With this in mind, I am inviting any website designers who might be interested to PM me. But before you all rush, please read the design brief I have prepared here.

It's probably not very clear or well written, but the gist is a website that will be a "one-stop shop" for all Albion historical enquiries - by underlaying a Wikipedia-style encyclopaedia (but with only authorised contributors) with a statistical database. It strikes me that this might be quite difficult or involved, so it will take someone with experience or someone with knowledge looking to extend their experience.

Note that the task for the designer is not to generate or even upload content, but to provide the template and means to allow authorised contributors to do so. We have plenty of content already: the database, the Albion A-Z and Seagulls books, numerous articles from our BHACHS newsletters, etc, etc.

Note, too, that there's not a huge amount of money to pay a salary!

It may be that there are better solutions than the one I have described, so I'm open to helpful suggestions as well regarding the design brief. The main part is the encyclopaedia-cum-database; other parts, like the on-line quiz, can wait. I envisage a timescale of around a year for completion.

It will be committing but, I hope, rewarding. But I would emphasise the need for discipline - sticking to the brief, putting design solutions onto paper, documenting the software, backing up software - in case, for whatever reason, the appointed designer is unable to see the project through to the end. In any case I hope this website will continue for many years to come, so proper documentation is essential for future maintenance.

If you are interested in this challenging but exciting and rewarding project then please PM me with your thoughts on the design, your relative experience, and any other information you think is relevant.

Thanks for reading this.

Tim Carder
Chairman, Brighton & Hove Albion Collectors' and Historians' Society

 

[Couldn't Be Hyypia]

All the best with this Tim. A year ago I would have offered you the resources of my team, but times being what they are...I no longer have a team! But I'm sure there are coders on here who could help you make a superb site. Can't wait - much needed resource.

 

[The Large One]

Have just forwarded this to a website designer - and Albion fan, of course.

 

[Superphil]

Good luck with this Tim, it will be the work of genius to put together this site, fingers crossed a suitable volunteer comes forward.

 

[Gary Leeds]

Looking through the brief I can tell you its going to be one hell of a database. Im assuming that for every single game played you will be listing the full team so thats going to be a huge database before you even start adding the rest of the information. It may be worth having a look at the wikipedia back end software at MediaWiki and seeing if that could be used out of the box so to speak. It looks on a brief check that it could cover everything you want to do but I dont know how it could handle queries like "all games played at Fratton Park" without you having to manually enter that information. The software could probably be changed to do this query but I doubt it would be a quick job.

I doubt I would have time to do the site the justice and time needed to set it up properly but would be willing to help out if required. I know a fair bit of PHP and SQL and will take a quick look at the wiki software tomorrow just to get a rough idea what would be involved to modify that. Possibly it would be worth starting with the default software as a base and then make the mods one at a time once most of the information has been added

 

[Cadiz Seagull]

Have just forwarded this to a website designer - and Albion fan, of course.

Tim, if it's the guy I think it is, I can thoroughly recommend him!

 

[The History Man]

Sorry, my apologies for the delay in replying to this thread, but I was suddenly diverted to other matters last week.

First the bad news. I have had no offers thus far! I thought it might be daunting. Perhaps it is too daunting.

Although I have used computers for many years and have programmed them (and embedded systems) in a multitude of languages, I have very little knowledge of websites beyond a little bit of HTML.

The database, as mentioned in the design brief, is in Microdoft Access. Although I'm sure it's more complicated in practice, in essence is it not possible to upload this database and use an SQL query to derive "All matches played at Fratton Park"? It's dead easy in Access. (A simple Yes/No answer, if possible, will suffice!)

Thanks for the interest. No one has actually commented adversely on what I am proposing, but if you do have sensible suggestions then please do let me know. Or a way forward!

 

[nail-Z]

Yes, it is possible to upload an access database to a web server and use it to drive queries through a web page. Not the most efficient method, but it is possible.

 

[Horsham Gull]

better off converting it to sql then uploading it

 

[clapham_gull]

Don't have a lot of time on my hands, what with posting on here. But in terms of database design, some advice:

I've done quite a bit of work (academically) on the structure of data that sits behind search engines.

If in particular there is descriptive information, it's well worth creating a few auxillary tables that index the individual terms. Rather than searching in a large text field.

You still retain than, but create a separate area to do the searching.

Most effective if the data is generally quite static.

With a design like that, it's easy to apply some simply "maths" into the queries that enable effective ranking of the results.

It's all about ranking the importance of certain words, and computing the probability.

So - what you get back is not so much, a load of results that are "true", more like a load of results than are ordered by truthfulness...

.. or "ranking" being the correct terminology.



You don't a specialised database to do it, just need to know how to do it.

If anyone is interested, google things like Inverted Document Index, Term Frequency and Inverted Document Frequency.

If that's hasn't sent you to sleep - look up the Vector Space Model....

A bit of work at the start like that makes the difference between (sorry Bozza) something like the search facilities on here against google.

1&1 Internet have cheap packages behind which you have a mySQL / SQL Server db, which is going to be a hell lot better than Access.

Looking at the brief, it would work particular well against the descriptive text of photos.