About a month ago, I got some old computer parts from Kah Wai which weren’t in use and set up a headless server using the latest Ubuntu server edition. I stumbled upon this guide which was excellent in guiding me through the process of setting it up and I would highly recommend it if you are new to setting up a server from scratch.

I’m pretty impressed by the Debian derived package management system which makes it very easy to add, remove or update almost any application available for it. I remember the times when I had to deal with all sorts of dependency errors when attempting to install stuff on Mandrake 9.0 (now called Mandriva). Besides the server services such as Apache, PHP, MySQL, I got Samba installed as well so it’s now also functioning as a dump box besides its primary function as a development server (not much so recently though..). :D

Pretty true to a certain extent. :P

via digg

Coincidentally, I found another pic lying on my hard disk of a similar vein:

I do not wish to prove this true during the usability testing session with my project’s client in 2 days time though.. It’s going to be a pain if that happens. XD

During my last trip to Mt. Hotham, I discovered that the videos I took with my IXUS 55 digital camera took up fairly substantial real estate on my 512MB SD card. For example, a 12 second clip took 21.9MB which when compared to videos available on the net these days, is extremely bloated (e.g. a TV show of about 40 minutes in length is about 350MB).

Checking the user manual of the camera reveals that it uses a Motion JPEG format for the video. This format stores each frame of the video separately which results in the much larger overall file size. Here are some results from the conversions which I did with the clips I have using VirtualDub‘s default XviD settings:

Before:

12 seconds, 21.9MB

1 min 3 seconds, 84.2MB

50 seconds, 57.3MB

7 seconds, 13.5MB

48 seconds, 75.0MB

After:

12 seconds, 1.75MB

1 min 3 seconds, 6.08MB

50 seconds, 3.67MB

7 seconds, 0.95MB

48 seconds, 5.61MB

As you can see, the results show quite a large reduction in size. Before following the steps below, you would need to ensure you have the XviD codec in your system. I’m using the K-lite mega codec pack which bundles all the various codecs (including XviD), but you can also download XviD standalone here. Steps:

1. Download the latest VirtualDub from here.
2. Extract the zip file that you have downloaded to your location of choice on your hard disk (e.g. C:\Program Files\VirtualDub).
3. Run VirtualDub.exe from the directory you extracted to (e.g. C:\Program Files\VirtualDub\VirtualDub.exe).
4. Open the video file you wish to convert using File->Open File or just drag and drop the file into the VirtualDub window.
5. Click ‘Video’->’Full processing mode’ (ensure there’s a dot on the left of Full processing mode).
6. Click ‘Video’->’Compression’.
7. Select ‘XviD MPEG4 Codec’, then click OK.
8. Click ‘File’->’Save as AVI..’
9. Choose the directory you want to save the converted video to, enter a filename and click ‘Save’.
10. The transcoding should now be in progress. The time taken for it to complete would depend on the speed of your processor and the length of the video.

It is possible to get even further file size reductions by tweaking configuration settings to sacrifice on quality. This usually requires some time in experimenting with various settings to get the best tradeoff which you want. You can read more about transcoding/encoding over here.

Do note that different camera manufacturers may use a different format for capturing video (for e.g. Quicktime MOV) which may already be optimised and would not attain the same amount of reduction in file size. ;)

As BitTorrent checks the integrity of every piece of data it receives, some parties have resorted to ‘swarm poisoning’ popular torrents which attempts to flood the peers with bad pieces of data which slows down the propagation of pieces as the client will have to re-download the pieces which failed the verification (hash checking). This leads to heaps of wasted bandwidth which is quite precious in some places like Australia where download quotas are enforced on most broadband plans. ;)

The following are the steps for Azureus and µTorrent. I believe these are the best 2 BT clients at the moment which you switch to immediately if you aren’t using one of them already. :P

Azureus (v2.4.0.2):

1. Start Azureus.
2. Click ‘Plugins’->’Installation Wizard’
3. Select ‘By list from sourceforge.net’, click ‘Next’
4. Scroll down to ‘Safepeer’ and check the checkbox on the left of it, click ‘Next’
5. Select ‘Install plugin(s) for all users’, click ‘Finish’
6. Click ‘Install’ in the dialog box that appears
7. Once download and installation is done, you can close all dialog boxes and remove the downloaded zip file from your seeding list

µTorrent (v1.6):

1. In µTorrent, go to ‘Options’->’Preferences’->’Advanced’. Change the ‘ipfilter.enable’ value to ‘True’.
2. Go to this site, scroll down the page till you reach the ‘Other Lists’ column header on the left navigation bar. Look for ‘ipfilter.dat.gz‘. Download that file to your PC (right-click->save target/link as).
3. Extract the file your archive utility of choice such as WinRAR. A file with the name ‘nipfilter.dat’ should be extracted.
4. Rename that file to ‘ipfilter.dat’.
5. Copy that file to ‘C:\Documents and Settings\your-username\Application Data\uTorrent’. If that file already exists, overwrite it.

Azureus’ Safepeer plugin automatically updates itself with the latest but you would have to do the steps above on a periodic basis for µTorrent if you wish to keep that filter file up-to-date. It’s probably not as critical as ensuring your virus definitions for your antivirus though, so you don’t have to do this on an extremely regular basis. :)

Edit: Updated ipfilter.dat links to TBG blocklists instead of BISS

_Edit (2010-09-19): To those of you who are coming via the Google search of a Bloglines replacement, I’ve actually switched to Google Reader nearly 4 years ago (October 2006) and have not looked back since. This article was more of a comparison of what was available in August 2006, and if you are looking for a Bloglines replacement in 2010, Google Reader is definitely what I would recommend switching to. :)

_

—

After heaping praises on Bloglines in my previous post on RSS feeds, it seems to have gone through quite a fair bit of downtime which wasn’t particularly pleasing. :( Feeling disgruntled, I started researching (again) on the latest breed of web based and desktop based RSS (feed) readers currently available to find a worthy replacement of Bloglines.

By checking out the features as well as screenshots from the respective RSS readers’ websites, I filtered down the readers (aggregators) which I would be testing more thoroughly to:

• Feedreader (desktop)
• Greatnews (desktop)
• Newsgator (web)
• Pluck (desktop + web)
• SharpReader (desktop)

Email applications which have RSS feed management:

• Mozilla Thunderbird (desktop)
• Outlook 2007 Beta 2 (desktop)

Notable absentees from that list would be Google Reader, Firefox‘s live bookmarks. As stated in my previous post, I have tried Google Reader and I loath the AJAX interface for it as it just doesn’t work as efficiently as Bloglines’s two pane interface. Neither do I like Firefox’s current live bookmarks setup.

To ensure all the RSS readers are tested in the same manner I would usually use Bloglines, I exported the whole list of feeds from it (which came up to about 93..). You can do this by clicking on the ‘Edit’ link just under the ‘My Blog’ tab when you’re viewing the ‘My Feeds’ tab on the left pane, and then click ‘Export Subscriptions’ link which is right at the bottom of that pane (under the header ‘Extras’).

Read on for the details..

Outlook 2007 Beta 2

{.imagelink}My software engineering project teammate was sort of raving about it and I was playing around with its calendars and reminders which lead to it being the first to be tested. Importing the exported feeds from Bloglines was relatively straightforward. But the moment that was done, the quirks begun to surface. Firstly, I didn’t quite like how the feeds were treated as separate email messages. It probably might be possible to get around this, but this leads to my third point which is there are way too many options which one has to go through just to get things to display accordingly. Secondly, it was awfully slow in synchronising with all the feeds after the initial run.

No doubt Outlook is an excellent email application as I’ve used it before during my internship but I don’t think it’s suited for those looking for an easy to use and configure RSS reader. It might appeal to those who already use Outlook for email/calendaring heavily, but not for the masses who rely on web based email.

Mozilla Thunderbird

Mozilla Thunderbird does the same thing as Outlook with message viewing but is a bit faster in synchronising with all the feeds. Another odd thing Outlook and Thunderbird show during the import is the loss of categories which I had in Bloglines.

.

Feedreader

Next is probably the most short lived test, which was Feedreader. After installation, it had 3 folders with 4-6 feeds in each. I couldn’t remove the folders straightaway as it pops up an error message saying that there are feeds in it. I eventually managed to remove all the default subscriptions by deleting off subscriptions.xml and headlines.xml from its application data directory, but the folders still required manually deleting. This default behaviour was deemed impractical to me, so I just moved on. :D

SharpReader

SharpReader synchronises with all the feeds pretty quickly. However, it uses the same interface like the others of separating each feed’s entries into separate messages.

.

Pluck

Pluck Firefox edition was tested next. It basically is a Firefox extension which uses a sidebar and also the status bar to notify of new updates. So it’s sort of a hybrid RSS reader being half desktop and web based. Overall, it seems pretty decent, displays feed entries sequentially without having to click through one by one. Unfortunately during testing, there were noticeable slowdowns at times when loading certain feeds, and certain feeds had entries which overflowed and overlapped the next one (might not be Pluck’s fault). Besides that, being a Firefox extension, this probably increases the memory usage of Firefox even further, which probably is something I would want to avoid.

I tried out the web edition of Pluck as well which is supposed to work like Bloglines. It is very very slow and had very bad usability which might explain why this edition is so well hidden on Pluck’s website (I had to use Google to search for it XD).

Best Two RSS readers

Now for the last two RSS readers which I believe are the closest in being able to replace Bloglines: Greatnews and Newsgator.

Greatnews

Greatnews was recommended to me by Khim Hoe as he uses it on a daily basis. First thing that surprised me was the blistering speed it had when doing an update on all the feeds especially when compared to the other desktop based RSS readers. Next plus about it was it did not render the feeds as separated email messages by default. And lastly, it uses CSS to determine how your feeds are displayed which can be tweaked to your hearts content. It even has a function to synchronize and retrieve feeds from Bloglines, but this would be pointless if Bloglines were to go down.

The only issue I had with it was the integration with IE as the rendering engine as I’m not able to use the extensions I have in Firefox which I am accustomed to as clicking on the feed titles or external links would open up as a separate tab in Greatnews itself. I tried changing the setting to open up links in Firefox instead, but having to go back and forth between the 2 applications got quite annoying over time (as opposed to Bloglines which would be just a switch between tabs). Its greatness was somewhat marred as I discovered a bug with feeds which use FeedBurner within an hour of using it (which is fixed in the latest nightly build).

Newsgator

Newsgator on the other hand is very similar to Bloglines as it is a web based RSS reader as well. Basic functionality was at par with Bloglines, but I yet again I found some weaknesses. As opposed to Bloglines which have a two pane setup using frames, Newsgator has both the feeds and content on the same page, so every click on another feed would require the whole page to reload again which makes it sluggish. It didn’t help either that the page loading times were a little slow as well.

Summary & Conclusion

After going through all the alternative RSS readers I have tested, I have a few points to make. Desktop based RSS readers seem to be lagging behind web based ones in terms of the features as well as usability. Next, it needs to synchronise with each and every feed’s server which you subscribe to on an interval which you set which may take a while and use up bandwidth that could’ve been saved by using a web based service. Lastly, a web based service allows you to access all your feeds anywhere which I have pointed out in my previous post.

Now to conclude on the reader which I would readily switch to. My answer is, none of those tested. XD In my opinion, Bloglines has the best interface, featureset, speed and the ability to handle a lot of feeds. Newsgator comes close, but unfortunately not close enough to compel a switch. Greatnews is probably the most promising desktop based RSS reader I’ve tried, although there still might be bugs in it which need to be ironed out.

To create some form of redundancy in case Bloglines goes down, you could create a Newsgator account and import all your feeds from Bloglines which I have resolved to doing (they’ve posted an entry here with regards to the last interruption). :D Maybe things would’ve been better if it was Google who acquired Bloglines instead of Ask Jeeves? :P