i’ve installed e text editor on my desktop and trying it out for some Ruby development.

E is heavily dependent on using Ruby in the background to perform a number of its refactoring and text-based actions. Unfortunately, sometimes you can end up in a situation where E reports the following error whenever you trigger one of these actions:

ruby: no such file to load — ubygems (LoadError)

It looks obvious that the error is caused because E is trying to load the gem ubygems when it actually meant rubygems, but the problem is more involved than just that.

In my case, it turns out that after running the Ruby one-click installer, the installation creates an system-level environment variable RUBYOPT=-rubygems. This variable setting is what’s giving E its grief.

To fix: Simply remove the environment variable from your system settings and restart e for safe measure. In my experience, there’s no need to download the rubygem package and re-install it as mentioned on another site.

I had the following email exchange this morning:

---------- Forwarded message ----------
From: Xerxes Battiwalla
Date: Wed, May 20, 2009 at 10:31 AM
Subject: Re: Email Account Verification Update
To: account_upgrade1@programmer.net


Sure thing - happy to oblige.

My username is: Whydontyou
And my pass is: Shoveitupyourarse

Enjoy!


On Wed, May 20, 2009 at 9:52 AM, HELPDESK ADMINISTRATOR
<Amanda.Warren@selu.edu> wrote:
>
> The Helpdesk Program that periodically checks the size of your e-mail
> space is sending you this information. The program runs weekly to ensure
> your inbox does not grow too large, thus preventing you from receiving or
> sending new e-mail. As this message is being sent, you have 18 megabytes
> (MB) or more stored in your inbox. To help us reset
> your space in our database, please enter your current
> username (_________________) password (_______________).
>
> You will receive a periodic alert if your inbox size is between 18 and 20
> MB. If your inbox size is 20 MB, a program on your Webmail will move your
> oldest e-mails to a folder in your home directory to ensure you can
> continue receiving incoming e-mail. You will be notified once this has
> taken place.
> If your inbox grows to 25 MB, you will be unable to receive new e-mail and
> it will be returned to sender. All this is programmed to ensure your
> e-mail continues to function well.
> Thank you for your cooperation.
>
> HELPDESK ADMINISTRATOR.
>
>
>
>
>
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
>


Just quietly, i’m surprised it managed to get through GMail’s spam filter….

In Windows 2000/XP you can add/remove content from your Send-To menu just by navigating to the SendTo directory of your profile (eg: C:Documents and SettingsXerxesSendTo

In Vista they changed the location of the SendTo directory and I never found out where until now! If you navigate to the shell extension “sendto” (eg: Start –> Run –> “shell:sendto”), it opens up the list of items in the send-to menu when you right-click a file in Explorer.

I normally like to add a shortcut Notepad so no matter what file type i’m opening, I can have quick access to drop it in Notepad if necessary.

I’ve used a lot of SCM‘s in my time, and none of them have been as esoteric as Git. This post serves as a reminder of the different ways to “revert” changes to a git repository

Scenario:

We are in directory with a local git repository. This repo contains 4 files and no sub-directories. Each file is in one of the 4 different states a file could be in for Git (not considering ignored files for the time being)

File Name State
Unchanged.txt File is unchanged in local directory
New.txt File is new to the repo
Deleted.txt File has been deleted from local directory
Modified.txt File has been modified in the local directory

All files except New.txt are being tracked and none of these changes have been staged/committed (yet).

Action: checkout

$> git checkout
  • This command has an implicit head of [HEAD]
  • This command has an implicit working file/directory of the CWD
  • This command will only affect tracked files
File Name State
Unchanged.txt no action
New.txt no action (because it’s untracked)
Deleted.txt no action (will not restore the file UNLESS explicitly stating file in command)
Modified.txt no action (will not restore the file UNLESS explicitly stating file in command)

Action: checkout <file/path>

$> git checkout .
$> git checkout Deleted.txt
  • This command has an implicit head of [HEAD]
  • This command has an explicit working file/directory of the path on line 1, and of the file Deleted.txt on line 2
  • This command will only affect tracked files
File Name State
Unchanged.txt no action
New.txt no action (because it’s untracked)
Deleted.txt File is reverted back to its original from the current head
Modified.txt File is reverted back to its original from the current head

Action: reset

$> git reset .
  • This command has an implicit head of [HEAD]
  • This command has an explicit working directory of the CWD
  • This command will unstage staged changes
File Name State
Unchanged.txt no action (changes are unstaged)
New.txt no action (changes are unstaged)
Deleted.txt no action (changes are unstaged)
Modified.txt no action (changes are unstaged)

Action: reset (with staged content)

$> git add .
$> git reset
  • This action assumes that all changes have been staged (line 1), so the repo is in the following state:
    File Name State
    Unchanged.txt File is unchanged in repo
    New.txt File is staged for adding
    Deleted.txt File has been staged for deletion
    Modified.txt File has been staged with modification
  • This command has an implicit head of [HEAD]
  • This command has an implicit working directory of the CWD
  • This command will unstage staged changes
File Name State
Unchanged.txt no action (changes are unstaged)
New.txt no action (changes are unstaged)
Deleted.txt no action (changes are unstaged)
Modified.txt no action (changes are unstaged)

Action: clean

$> git clean -d -f
  • This command has an implicit head of [HEAD]
  • This command has an explicit working directory of the CWD
  • This command will remove files and directories which are untracked in the repo
File Name State
Unchanged.txt no action
New.txt File would be deleted
Deleted.txt no action
Modified.txt no action

So in summary, if you want to completely revert your working directory to a clean state (IE: the equivalent of an svn revert) is to:

$> git clean -fd
$> git checkout .

I think it’s safe for me to say “shea right – if i hate AOL Search as much as I despise AOL the ISP, this article will not be favourable to AOL in any way, shape or form”.

Let the games begin.

AOL Search fails to render properly in Google Chrome

AOL Search fails to render properly in Google Chrome

Wow off to a flying start here, boys…. </sarcasm>

The second thing to peeve me off is that AOL Search doesn’t have a search provider exposed in their meta-data. So I have to create one for myself. Fortunately, Chrome makes this pretty easy, but that’s not the point – You guys are providing a search service. Irrespective of how shit it may or may not be, FFS at least make it easy for me to *TRY* and use your product?

…all this, and I haven’t actually started using it yet. How ominous. I’m hoping that the little “powered by Google” actually means “we cant do search anymore, and have given up. Here’s one which does it better”.

……………

One week (maybe a little more) has passed and well, lets just say i’m not as disappointed as I thought I would be. Mainly because AOL search really does seem to be effectively a wrapper around Google. As an example, I searched the hottest topic going around on the tubes at the moment (the effects of socialism on post-war Germany), and most of the results were the same, except that Google also listed a result to its Book Search service. Apart from little things like that, these two are inseparable. Even AOL’s image search is just a face-mask over Google Images.

The design of the page leaves a little to be desired however, as AOL shamelessly place advertisements on the top of the page in an attempt to drive click-throughs. My ad-busting eyeballs detect this easily so the placement of the ad isn’t so much the problem. The problem is that they have sneakily set the length of the HREF for each paid link to be the full width of the page, which means by clicking in what should be “blank space” you trigger the link and click-through the paid ad. Naughty, naughty.

All said and done, I couldn’t help but realise I just commented to a colleauge without realising that I’m finishing up this post so that “I can go back to using Google”. I guess even subconsciously I find any experience outside Google’s to be less than engaging.

On a final note, The Wolfram Alpha didn’t launch as soon as I was hoping it would, so there’ll be a week’s rest where I go back to Google, before trying out the new kid on the block on 18th May. Yes, I am aware of the broad misrepresentation of Alpha being as “Google killer” but it would still be fun to try :)

In this, the 3rd installment of The (Not So) Great Search Engine Showdown, I reflect on my experience using Ask.com compared to Google.

I don’t have a great deal of time so this post is going to be brief. I really only have one _serious_ gripe about Ask – that stupid fu#@$%g Answerbar at the top of the page everytime you navigate to a search result. NO, ask.com! I wanted you to give me the search result, not a pain in the ass waste of screen real estate. What also frustrated me about this “feature” was it’s sheer unpredictability. Most web results would display the Anusbar at the top, but others (like Wikipedia) would be displayed in full glory without being crippled.
The “Close Permanently” button was never hit with such gusto, i’m sure. To demonstrate just how much, i’ve prepared the following illustration:

How to close the Ask.com Answerbar

How to close the Ask.com Answerbar

By way of quality of results, I actually found Ask to be better than I was expecting. Certainly I felt like I wasn’t missing Google, though on a few occassions I had to drop back just to be sure I wasn’t missing anything (turns out I wasn’t). Overall the web search results were as good as Yahoo’s, though one thing that irritated was that Ask.com mixes the paid advertising results in with the organic search results. I’m sure they’ll claim that they’re putting the top-most organic result first and then allowing the rest of the results to be shown underneath the paid section, but we all know the truth. Money grubbers.

When it was originally launched as “Ask Jeeves”, the website’s search technology was based on doing some NLP against your search query and it would try to return the best results based on the context of your question. A few years ago Jeeve’s was given the arse from his job, and the company took the arse to their search results, because (quite simply) their NLP wasn’t advanced enough to provide accurate results compared to Big Brother

However having played with with Ask.com this week, I noticed they still have a Q&A section (it claims is in Beta) which allows you to phrase a question and let the NLP try and answer it for you. Not one to turn down a good opportunity to test NLP products (and get a comparative feeling for the upcoming Wolfram Alpha test i’ll hopefully be performing), I Ask’ed the following question in the name of science:

Putting Ask.com's NLP to the Public Service Announcement test.

Putting Ask.com's NLP to the Public Service Announcement test.

It’s heart-warming to see that even if you speak broken English like the second guy, you can still get valuable advice on the interwebs.

This week, I throw away all credibility as I try out AOL’s search. If using this website results in me getting another fking AOL starter CD, i’ll sht the roof.