May 31, 2005
It's getting weird
OK now it's getting weird. My entries on Destination is appearing on Misspelt and now the duplicate is on the Panic! post. I suspect some database corruption.
Posted by vhadiant at 10:41 PM | Comments (0) | TrackBack
May 30, 2005
Lost 2 post.
Crap I just lost 2 posts from Misspelt, had to recreate the damn thing again. This is getting really annoying :(
Posted by vhadiant at 10:49 PM | Comments (0) | TrackBack
Panic!
Whoops I'm in a bit of panic. Just now I lost all my blog entries. It started when I'm about to post on Sydney's public transport on Misspelt, it threw an error and suddenly all my entries dissapear, just like that!. On MovableType main menu, when I log in I can see my two blogs on the main menu, the total entries (which is correct) but I can't see the entries! Strange, I can see the list of comments and trackbacks but try as I might I can't coerce MT to show the entries.
After fooling around for a while somehow it reappeared, and my last entriy on Misspelt is duplicated. I tried to delete the duplicates but I could only delete 1 and the other 2 still there and at the bottom of MT I saw this bunch of error message:
substr outside of string at lib/MT/Serialize.pm line 58.
Use of uninitialized value at lib/MT/Serialize.pm line 58.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
substr outside of string at lib/MT/Serialize.pm line 58.
Use of uninitialized value at lib/MT/Serialize.pm line 58.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
Use of uninitialized value at lib/MT/App/CMS.pm line 2442.
Use of uninitialized value at lib/MT/Serialize.pm line 59.
substr outside of string at lib/MT/Serialize.pm line 59.
Use of uninitialized value at lib/MT/Serialize.pm line 60.
Use of uninitialized value at lib/MT/App/CMS.pm line 2442.
I tried to check what the heck happen on Serialize.pm 58, but I forgot that Perl is the only programming language that looks the same before and after source code encryption, grrh ... what's going on MT, I thought you're a stable and rock solid platform.
Bah it's late now, I'll have a look again tomorrow, and wait what people on the MT forum said about this problem. Going to click "Save" now, hoping it'll be OK.
Posted by vhadiant at 10:29 PM | Comments (0) | TrackBack
Google's ingredients
I have some interesting notes that I want to write down. For some reasons I’ve had a number of discussions with a couple of my friends about Google and how they become what they are.
1) Google pretty much solved the web search problem. Although the information retrieval space has been around long before Google, they’re the first to solve the web search. This is mainly because of their PageRank algorithm. Alhtough the information retrieval space has been around long before Google, they're the first one who nailed the web search problem. They’ve discovered that one of the most important properties of the web is the linking between the pages, and by factoring the links in their search algorithm they’ve managed to figure out which page is the most relevant to your query. Sounds simple isn’t it? But no one figured it out before Google.
2) Google figured out the long tail. So PageRank is probably the next best think since sliced bread, but without the money Google would have become another relic of the Internet.
3) Google appeals to the geeks and yet manage to not put off the non-geeks. Google is good, quick, simple, and no-fuzz. Their ads are discreet and doesn’t cause eye sore, geeks love Google. And who do people ask for advice for anything remotely related to technology? The geeks of course! And Google is one of the few things that geeks are not afraid to recommend to others.
4) Google is smart, and does not stop their innovation engine. Although their latest feature to be able to personlise Google’s homepage puzzle me a bit, it just doesn’t seem like Google. Regardless, there’s some really uber cool stuff from the labs, the maps, citation, definition, groups etc.
Anyway, I’m sure there are tons of other articles on the net on similar issue :)
Posted by vhadiant at 06:47 PM | Comments (1)
May 23, 2005
100 GB Blue Ray disc
100 GB disc holy cow ....
Posted by vhadiant at 07:22 PM | Comments (0) | TrackBack
I Google, is it the end?
Last week Google launched the "home page customisation" feature on the labs. It links to google.com/ig. Not sure what ig stands for, I call it "I Google".
It is pretty much a portal and the design curiously very similar to Yahoo portal. Tell you what, I'm not really surprised anymore after GoogleGroups and GMail. Google has moved away from the simple come in -> do you quick work -> leave metaphor to "come in and stay". Is it the end for Google? With all the other innovations out from their labs (map, citation etc) why would they put this out? Have they run out of new innovations? Hopefully not ...
Posted by vhadiant at 07:15 PM | Comments (0) | TrackBack
May 19, 2005
Useful tool: CruiseControl
I've been using CruiseControl for the past week and half, and I found it extremely useful tool, in fact put it under the "Must Install" component in any project you're working with. It does a number of stuff, but the most useful bit is the ability to monitor any commit on your source control, automatically grab the latest version, build it and run all your unit test. Since we're working on a .NET project we're using CruiseControl.NET.
All the developers would need to install a CC.NET sys tray app which polls the status of the CruiseControl.NET server every couple of seconds. When someone committed a bad stuff, it goes red. Double click on the sys tray icon and it opens the last build log and shows who committed the bodged files. Pretty cool huh, I only wish that instead of displaying "Last check-in has broken the build" (when there's something wrong) it says "XXX has broken the build, FIX IT!". On second though that's probably can be configured on the server, but I don't have access to our source control server heh ...
Posted by vhadiant at 07:24 AM | Comments (0) | TrackBack
May 17, 2005
NewsGator Acquires FeedDemon
So NewsGator acquires FeedDemon, interesting move by both company, and somehow after reading Nick's explanation on his blog I don't buy it. Both are RSS readers, both claim that they do more than just RSS reading but it's pretty much what those two software do, in fact FeedDemon is a direct competitor of NewsGator.
There's a lot of congratulation notes on Nick's blog post, but I suspect the general response will be "What the ...". On another note, I've been proven to be wrong ... many times :)
Posted by vhadiant at 10:24 PM | Comments (0) | TrackBack
Annoying Visual Studio.Net error
Grrrgghh, for the past few days I've been plagued with the "The process cannot access the file because it is being used by another process" error during build. This often happen to your Visual Studio.Net solution if you have multiple assemblies in your solution as described here:
http://support.microsoft.com/default.aspx?scid=kb;en-us;313512
Whatever it is, I can't use the workaround described in the aforementioned KB. So to fix this just close your solution, go to the offending project, delete the debug directory in both Bin and Obj directory and then rebuild. If this doesn't work, you have to close Visual Studio.Net itself. If that still doesn't work, check that the problem assembly do not have a reference to itself, although the offending assembly will compile with no issue the other assembly that refer to this one may complain.
Posted by vhadiant at 10:08 PM | Comments (0) | TrackBack
May 15, 2005
Blogging is a fad?
Scoble retorts to a report that said very few people are reading blogs. The "anti-blog-hype-backlash" is coming he reckons. I agree that blogging hype is not over, but I think Scoble need to swallow the bitter pill of truth that in fact there are very few people who read blogs.
Take a random sample people, how many do you think have heard the word "blog"? Let alone understand what the concept is about? On my own group of friends (which most work in the IT field), ex-co workers (barring Synop of course) and current co-workers, I am surprised at the number of people who actually know what a blog is, a few indeed. He is taking an example of his own blog on how quick you can spread the words, that’s a bit unfair :) he is an a-list blogger!
I'm not saying that's the blogging phenomenon is slowing down, but right at the moment it just doesn’t have number yet.
Posted by vhadiant at 08:43 PM | Comments (0) | TrackBack
May 05, 2005
Gmail slowing down
I notice that Gmail now is slowing down the size count. It still increasing (currently at 2163.309631) but at a lot slower rate than before.
Posted by vhadiant at 10:08 PM | Comments (0) | TrackBack
May 03, 2005
When one becomes two
I've decided to split my blog into two. One for personal stuff and this one for tech stuff. Not entirely sure yet if this is going to work better or worse, but I do get annoyed every now and then reading other technical blogs with a couple of personal posts here and there.
It does add more maintenance effort though, I have two blogs with (slightly) different templates and style to manage. I think I'll make them similar with different colour to identify the two different blogs. Maybe when I have more time in the future I'll do more work on the layout.
Posted by vhadiant at 11:10 PM | Comments (1) | TrackBack
