Dented Reality

All posts tagged 'search'

Twitter’s Missing Features

I’ve been sketching some random thoughts and plans for a new Twitter client, and rather than restrict it to my notebooks and condemn it to the fate of so many other ideas of mine that never see realization, I’m putting my wish-list for what a decent Twitter client should allow me to do here. Feel free to roll these into another client and let me know so that I can just use it, rather than have to build it :)

Read the full list after the break

Twitter Search Bookmarklet

Throughout the day, I’m likely to search Twitter a number of times. Being the lazy person that I am, I got sick of opening a new tab, typing search.twitter.com, hitting enter, waiting for the page, then typing my term and waiting for results. I already have a couple of other bookmarklets that I use all the time (one for loading up a Google Map, one for jumping to a PHP function name in the online reference), so I figured I’d knock up something to search Twitter. I’m using Firefox on a Mac, YMMV.

  1. Bookmark any page (seriously, any page).
  2. Put that bookmark into your Bookmarks Toolbar.
  3. Right click it and select Properties
  4. Change the Name to something like “Search Twitter”
  5. Change the Location to this (copy and paste it):
    javascript:var%20d=document,w=window,e=w.getSelection,k=d.getSelection,x=d.selection,s=(e?e():(k)?k():(x?x.createRange().text:0));if(!s.length){s=prompt('Search%20Twitter:','');}if(s){document.location.href='http://search.twitter.com/search?q='+escape(s);}else{void(0);}
  6. Save Changes

That’s it. Now click that link in your Bookmark Toolbar and it’ll prompt you for a search string. Enter anything and hit OK. If you happen to have some text on the page selected, it’ll automatically search for that text. Voila – Twitter Search.

First Notes On My Search System

View an Acrobat PDF document (13.6Kb) with more information regarding the search system

I wanted to get some information out here so that people could see what I am working on, and indeed what is running this site’s search facility. The icon to the right will take you to a PDF document (13.6Kb) which gives some more details about the approximate process flow of the search supplements, as well as a database schema of what’s running the best bets, variants and related terms.

Another O’Reilly Article On The Table!

I’ve been emailed by the author of an upcoming O’Reilly article for techo’s regarding my work with a more complete search system. He wants to discuss my integration of Best Bets, a controlled vocabulary and a normal, keyword-based search engine. There’s not really a lot to say right now, but I’ve emailed him back saying that I am definitely interested, so we’ll see what happens.

Interest From SearchLoggers Group

I’ve had some initial interest from some of the other members of the searchloggers group in the work that I am doing with the search system on this site. The integration of a controlled vocabulary, best bets and a stand-aloone search engine backend appears to be of interest to others as well :)

My hope is that this can be developed into a stand-alone product that could be wrapped around nearly any search system, simply adding to the functionality of said search system. So far things are looking good, but only really for something written in PHP. I would need to add other options for systems written in Perl, C etc.

In the near future, I will be posting database schemas, some notes on implementation and other bits and pieces about how I am attacking the whole process. Keep an eye on things if you are interested :)

Trudging Along

Work on my site is coming along. I have added in blosxom v1.1, which went pretty much without a hitch. I had to modify my template slightly, but that was more because of my dodgy custom handling, rather than anything to do with blosxom itself.

I’ve been spending some time on the projects section, getting the project pages up to scratch and re-formatted; it’s looking pretty good. I am taking a few of the old projects offline, because they don’t have any documentation, don’t work anymore, things like that :)

I’m not going to put a date on when the site will be live, but given the current progress, and the list of things to do still, I would estimate about 2 weeks. Here’s the current list;

  1. Projects Sections
    1. Blogger API (functions, classes + meta)
    2. phpMassMail (also requires some work to make it 4.2+ compatible, and re-format documentation to include)
    3. JSSearch (would like to get some more information/documentation/examples included)
    4. JSValidate (want to update this to include the ability to open a popup window rather than an alert – optional)
    5. Client work and websites
  2. Contact Form
    1. Layout/design
    2. Contingency design
    3. Processing/handling
    4. Result/output page
  3. Search System
    1. Layout/Design (of the results, as well as extra options, defining manual entries etc)
    2. Processing system (integrating my manual results with Google/XooMLe’s results)
    3. Management of my manual entries (and a decent name for them :)

So there we go. That’s what I’ll be working on in the near future. And for those interested, I will be creating a custom search system, which integrates Google’s results for within my website (using XooMLe) with a selection of manual “Top Picks” or “Best Bets” which I have selected for certain terms. I have the rough idea planned out already, just need to implement it in code. I will probably make the code available for download once it’s complete as well, so keep an eye out for that :)

That’s all for now – time to get back to life.