I was trying out the new Bing search engine from Microsoft for various searches over the last few days (for things I would normally use Google) just to get a feel for how well it performs. Like almost everyone else, I found the results were surprisingly good — comparable to results from Google. But as I was using it, I noticed that one of the reasons I found the Bing results good perhaps (now this it’s very subjective) is because they present the Bing search results in the same format as Google search —
they seem to be using the same font with the link text in blue, 2 or 3 lines of description in black and the actual URL in green below the description.
they have a link to a cached page
they highlight the searched terms in the above summary
even the sizes of the font seem to be similar
My theory is that Microsoft conspired to copy the result format to help people psychologically accept the results of being just as good as Google’s!
See a couple of screen shots below:
Compare the results from Google above with the results from Bing below.
Create a file with the above contents (make sure to join lines that have been split), name it bing.xml, stick it into C:\Program Files\Mozilla Firefox\searchplugins and restart Firefox — now you have bing in your search list.
on doing “package require tdom“, the problem is related to the TCL implementation being multi-threaded while the version of tdom library that is being loaded was built without multi-threading support. The fix is to install a version of tdom with multi-threading support. Run “teacup install tdom” to get latest from ActiveState. It seems like version 0.8.2+ of the Windows DLL of tdom is built with multithreading support. It is possible that you already have multiple versions of tdom. So try using the package command with the version option as “package require tdom 0.8.2“.
Microsoft Windows Desktop Search is a decent tool to index all your files and emails and search them. Unfortunately the user interface is too minimal lacking even a help menu that allows you check on syntax etc. So here below is some info on the advance search syntax that you can use with Microsoft Windows Desktop Search. My explanation is focused on searching email since I use Microsoft Windows Desktop Search primarily in conjunction with Outlook.
You can combine search terms using the keywords “NOT” and “OR” (use in upper-case, default is to “and”). You can’t combine multiple keywords in the same query. Use parentheses to group words that need to be separated by the keyword (NOR, OR).
Look for all sources containing the words “one” and “thing” (case and order insensitive)
one NOT other
All sources containing “one”, but not “other” (case-insensitive)
one OR other
All sources containing “one”, “other” or both (case-insensitive)
All sources containing the exact phrase “one thing” (case-insensitive)
All sources containing (if ‘true’ specified) or not containing (if ‘false’ specified’) attachments. Can be combined with other search terms.
Sources that are (true) or are not (false) attachments. Use this in conjunction with a search term to only look inside the attachment, but not the source containing the attachment itself.
date:, before:, after:
Use to specify dates. You can use the following terms “yesterday”, “today”, “tomorrow”, “thisweek”, “nextweek”, “lastweek”, “thismonth”, “nextmonth”, “pastmonth”, “lastmonth”, “thisyear”, “lastyear”. You can also give names of months and names of days (in which case it refers to the current year or the past week). Also you can give dates in mm/dd/yyyy format. You can specify years by themself too. You can preceed the date with the greater-than (‘>’) or less-than (‘<‘) sign to specify ranges. You can include two dates to when preceeding the date with the greater-than or less-than signs. e.g. date:>1/1/2008<1/31/2008.
subject:, from:, cc:, bcc:, to:, received:, sent:
Use to search specific fields of an e-mail. The date type fields can use specifiers as for the “date:” syntax.
For some reason, a new filter that I added to my list of filters in Thunderbird was not getting applied (Thunderbird version 3.0 Beta 2) and I noticed that one of the buttons on the “Message Filters” dialog was a “Filter Log” button. So I went ahead and clicked that hoping I could seem some log of filter activity. Unfortunately this just brings up a dialog that allows you to turn the filter log on or off and to clear it. The default was to not maintain the filter log, so I turned the log on. Then I tried re-running the filters and nothing happened. So I thought I would look in the filter log file — but wait, where the heck is the Mozilla Thunderbird Filter Log File? Note that I am using Thunderbird on Windows XP. Hmm. Searching through all the menu options did not point me to anything that would show me the log file. Then I surmised that this “filter log” must be a dev type feature (either that or they still haven’t gotten round to developing this feature fully yet) and did a Google search — sure enough there were many people talking about the filter log file, but very few actually mention where the Thunderbird Filter Log File is on Windows. Some more digging and going in and out of the various links and I find that…ta-da…the Mozilla Thunderbird Filter Log file is named “filterlog.html”, but still no mention of where the file could be. Rather than visit more web-sites in search of the location, I decided to search all the folders on my disk for this file and…..didn’t find this file! Hmm. This is getting curiouser. It seems like Thunderbird 3.0 Beta 2 is really beta quality software. The finally I found a page on the Mozillazine web-site that indicated that the filterlog.html file should be in the Profile directory and that the Profile directory would be in the %APPDATA% directory which essentially on XP is “C:\Document and Settings\<Windows user name>\Application Data\Thunderbird\Profiles\<Profile name>. I looked there and did not find the “filterlog.html”. Seems like a bug or issue or perhaps it’s late and I am not thinking this through right. I give up…if you know what I am doing wrong, just email me. Thanks.
And why the hell is a log file in HTML format? It makes it hard to process and search for stuff using an editor….
Woohoo! My site experienced the Reddit Effect first-hand. Since the visits were to a static-page, there was no trouble at all.
Typically my site, http://grok2.com, receives about 20-30 visits a day. Mostly from organic Google search results.
Then, some time on Saturday 12th, April 2009, someone posted a link to the page “Why is programming fun?” on proggit (http://programming.reddit.com).
And soon the visits started piling up. April 13th saw the full brunt of the Reddit Effect with visits from Reddit alone of about 11,825. The visits were way down on April 14th, but still were pretty decent at about 1400 visits.
A couple of things are interesting to me. One is that the average time on the site was in the order of seconds. It is interesting that so many visitors didn’t really read the content on the site, but there were a reasonably large number of comments on Reddit — perhaps most visitors were quick readers of prose :-).
Another thing that was interesting was the side-effect of appearing on Reddit. It seems like Reddit links get referred to by users on places like Delicious and Hacker News resulting in additional visits from these places. I’ve included a pic showing the various other places I got a visit from below. As a result of these additional sources, on April 13th, the total visits peaked at 17,204 according to Google Analytics.
Overall it seems like someone took a fancy to the quote I had from Fred Brooks’ “The Mythical Man-Month” book and posted it on Reddit, but the other content on the site was not attractive enough to the visitors of the site. Need to work on that :-). But I am pleased that a page on the site was featured on Reddit. Thanks Reddit user pbkobold.