debugging my aunt, who is bugging me

So I get the usual near quaterly phone call from my aunt.

Her computer is getting slow and she wants a new one. She bought, not less than 4 years ago, a now discontinued Sony Vaio all-in-one Desktop (not too disimilar from this). It is beautiful of course, designed at the time to compete with Apple's increasingly beautiful and powerful iMacs. So it is no slouch, and can easily consume the tiny appitite of my aunts entire knowledge of applications (IE, Outlook, and Word).

The new PC which she insisted on taking me to see in the store is again a discontinued -- but still powerful -- LA series like machine, Vista installed (which she wanted removed and XP put on), beautiful -- of course -- and £1200! Not to mention way more powerful than she would ever need. Infront of the Sony shop staff, who she brilliantly orders around, I batted off his insane and wholy rehersed schpeil about how this is the right machine for her, and thwarted a few more of his rediculous answers to my aunts similarly insane queries about universal remote controls.

At this point, I guess I should point out that when my aunt, who is in her 60s, last took a course on computers about the time I was born (a little over 20 years). She even agrees with me that she doesn't understand how things work now-a-days, but still won't take on any advice. Many times I have tried to convince her to put her files in My Documents not 'C:\WPD50\', I guess there isn't much wrong with it, it's just another directory. But why put stuff there when My Documents is the default directory for any application and is accessible in 2 clicks.

Getting back to her house, I convinced her to see sense, and that her computer is perfectly sufficient. I put it bluntly:

"I would love to tell you to get rid of that old one, because I would happily take it off your hands, but there is absolutly no reason to take the current one away".

When I got on the computer, it didn't seem slow to me, but I knew I had to do some tinkering and look around to see what might be causing the perceived problem. Disk space is fine, around 40% used on C:, plus a second 120gb drive that is empty apart from crappy Sony branded applications, SonicStage anyone?

I was pretty stumped, I could defrag, but there was nothing to defrag. I searched for large files, a couple of sample videos from more crappy Sony applications. I then saw the Recycle Bin had something in it, so I clicked Empty... 20mins remaining... WTF... what is in there?

I canceled it and had a breathed in deeply before double clicking. It contained over 5,500 files, 50% word documents, 50% text documents ranging from 1k to 200k. I yelped, and swung round to my aunt on the other side of the room, and asked her how she had accumilated so much stuff. She went through how when she is done with an article she has written (shes a journalist), saves it, and then saves a text copy. She then goes in to Outlook, starts a new e-mail and goes through the menus to insert the text file, and deletes the copy. She claims this is because the news room will only accept text, and not word documents. I tried to explain to her the merrits of copy and paste, but she could not conceive how it would ignore the formatting, or as she explained it, the "funny character bits infront of the text" -- think older Word applications.

Consider that for a moment; lets assume she is putting formatting in her word document, she knows it is going to be removed when she e-mails it, So why does she do it? The alternative is that she doesn't put any formatting in -- which she doesn't anyway. She didn't get it.

So that acounts for 50% of the files, what about the other 2,500? She explains to me that when she no longer has a use for the file, she deletes it. So they are duplicates. Occasionally, however, the editor may ring up and request an old article, so she simply goes to the Recycle Bin, recovers it, and mails it again. I note that a lot of the files have the same name, she explains it as being 'catch words' so that when they go to the news room, they know what the e-mail is about (isn't that what the e-mail subject is for?). I realise at this point that when she 'no longer has a use' for these files, she actually means 2 files end up with the same name. In the Recycle Bin, there is no such naming restriction. Essentially, she is using the Recycle Bin as an archive with a loose naming structure. I also realise the news room has a retarded archiving policy, keeping the paper copys but not the digital ones.

So I set up an archiving policy for her. I clear out the spare 120gb drive, and dump everything in there, saying no to the write over, and creating a folder called older, to put the duplicate name files in, itirating the creating of the older folder until I have run out of duplicates. I perminently delete the text file duplicates. I also get her to promise to go on one of those European Computer Driving Licence (ECDL) courses, and drop old stuff on to that drive giving them a decent name. I at least manage to convince her that files can be more than 8 letters long, and contain spaces.

I guess the experience is educational, in some ways I wish I had a file system that allowed me to name things with all the same name if I so chose, it would make archiving easier. Also it made me realise quite how much people do things because they've always done them.

I also bet at no point in the ECDL course, my aunt will get told "Don't save stuff to the Recycle Bin".

A few days later she left a message on my voice mail praising me for my wonderful work, she even claimed that her keyboard and mouse now seemed to be responding better, but I have to go back next week to install a newer office than 2000 on her machine. I could completely blow her out of the water with 2007.

Bob Geldof and 82ASK

This is Mick Jagger. Mick Jagger is a puppet. In Hard Rock Cafe London.
Me and a few friends went to Hard Rock Cafe London last Thursday, and a good time was had by all. The biggest surprise of the evening was a guest appearance by African child saviour and possibly King of rock music, Bob Geldof. We later sang Happy Birthday to his daughter Twinkle, or Tinstel... or some... Tigerlilly, that's it. I have a friend with a cat called Tigerlilly too. What are the chances?

We then wanted to find out more about Sir Bob, and his daughters, more specifically, so we cranked out the phone and texted 82ASK. After a little discussion about how to phrase the question, we went for "Who are Bob Geldof's daughters?". We very quickly got a response, telling his about his 3 daughters and their ages. We immediately after got another text saying "Reply with the word HAPPY to get a free hour of questions." We didn't need to be told twice.

It was actually kind of difficult to think of questions to ask, we kicked off with "How tall is Bob Geldof?" answer? 6ft 1.5inches. We found the weirdest question they've even been asked "How tall is a Smurf?" answer? 3apples high. And the best joke "Why are Pirates, Pirates?" Answer? Because they ARRR! - Yeah, that was a bit shit, but it was free. I do wonder if they have stock answers to those questions, and it also makes me wonder how their system works. We did ask them how many people answer our questions, but they wouldn't tell us.

After all our questions, 12 or so, and our hour running out, we felt it was only fair to return the favour, so we texted them, letting them know where Bob Geldof was at that minute, their response? Nothing...
Very weird.

Oh, also, I wanted to add. With any luck, there shall be some cool updates in the near future.

How to download Flash videos from youTube, GoogleVideo, GameTrailers without the use of crappy tools like VideoDownloader

I would have made the title longer, if it would fit:

How to download Flash videos from youTube, Google Video, GameTrailers, or even PornoTube without the use of crappy tools like VideoDownloader which don't even work.

Now, after using all those crappy add-ons for Firefox that aren't actually any good, and very rarely work. I have finally found a solution that works every time... manual delving.

Thanks to this fantastic site from digg.com, which discusses the really simple side to Firefox that people don't think about, one of them was the 'about:cache' tool, which lists all of the items that Firefox currently knows about in memory and stored on the hard disk. Just type about:cache in your address bar and follow the link for each device.

Putting 2 and 2 together lead me to the realisation that one of these -- most likely on the disk -- knew the URL where the FLV files for flash containers would be stored. So load up the video, from GameTrailers.com refresh the cache page, do a quick search for .flv and there it is.

This should work for almost any video site, i've tested GameTrailers and PornoTube.com The only ones that are different are Google Video and youTube, where you're best bet is to search for the document ID, as follows:




The Google one doesn't come out so well because the URL is so damn, long but you get the idea. You might notice the file size is 0 bytes also, I'm not sure why this is, for the moment; I can only assume its a bug. You could of course search for the files by document ID, but .FLV will bring up less results and involve less searching.

So the steps are:

1. Go to the video from your favourite site
   
2. Load the about:cache?device=memory
3. Do a find for .flv or document ID (ctrl+f or /)
4. Copy the URL in to your download manager or click on the link and use the URL in there
   
5. Convert to your favourite format type

Hopefully someone will make an extension to do this sometime in the near future. But for now this is the best way of doing it without your add-on phoning home every time you want to download a file.

P.s. Let me know if it works for you're favourite site.

The Digg Commenting System Just Does Not Work

This issue is certainly something that has been covered before. But I am going to explain what my personal experiences and lessons are from this.
I think the problem probably stems from Digg's popularity, and by that, I mean there are a lot of idiots, and by that, I mean fanboys.

The difference seems to be that the people using Digg have a very wide ranging area of topics they like to view, but they don't view it in much detail. The wide but shallow approach. This is probably from how it works, pages that are popular are going to be popular to the widest range of people. But because the site aggregates everyone in to one site, people often comment without feeling to need to fully understand a subject.

What this means for the commenting system is that you end up with some very narrow minded individuals seeing a lot of things they don't agree with, along with the stuff they like and do agree with, and via some small minded mentality feel they need to put everyone in their place. And the simplest outlet for this is to digg down people.

Now I'm a person who likes feedback, if I've done wrong, what stops me from doing wrong again, if I have no feedback. And a yes/no voting system is not enough. Interestingly -- probably because yes/no voting is so easy -- you get very little comprehensible feedback from users, as you can see from my image above which was posted here, a hotbed for fanboy activity.

There are 2 solutions, one is to ignore blatant fanboy activity such like:

"I have about 12+ games for my DS, if I had my way, I would own more, Club House games provides infinite replay, but ignoring that, I havent yet finished final fantasy 3 or yoshis island... my problem with the DS is it has too many good games on it, can you say the same about the PSP?"

The other is to get the Digg team to add a simple 'opinion' feature, like Slashdot that lets you vote how funny/stupid/useful etc. the comment is.
Generally speaking I think the system of non-proportional representation within Digg is really letting it down, a system that gives higher vote to certain users would certainly be something worth looking in to.

Government E-Petition - Why Bother?

The UK Government's E-petition's website, on paper sounds like a great idea. It's a similar concept to PetitionOnline where users sign up and put their name on a petition that they feel strongly against. Usually these things are a waste of time, they carry no clout, anywhere. So I only add my name if it's something I feel really strongly against and I can see no other sensible recourse.

The Government's E-petition website is different though, set up and maintained by the government in partnership with mySociety, a charitable organisation "which builds websites that give people simple, tangible benefits in the civic and community aspects of their lives". Their plan seems very respectable, I take my hat off to them. But the government end of the site seems to have let them down.

I think it needs to be said, the site design, code, all that seems well done. But mySociety's mission statement "to teach the public and voluntary sectors... how to most efficiently use the internet to improve lives." seems to have missed the mark.

I personally have signed up for 2 petitions now, like ~90% of the country I put my name down on the "Scrap the planned vehicle tracking and road pricing policy" petition - which is closed - and more recently the "Change the current student loan interest repayment, to deduct payments monthly not annually" - still open.

But the responses I have got have been far from encouraging. I felt the responses that I got, far from alleviating my concerns, gave me greater anxiety for new reasons, simply because no one seems to be listening.

I would like to see some figures on the vehicle tracking scheme, but I bet you it is in the several 10,000s, if not much more. But much like the march on the war in Iraq, the government has virtually said 'We don't care, we know what's best, we're going to do it anyway'. And that's the bit that worries me, that bit is the difference between a democracy and a dictatorship.

It's probably a more of a problem with the system, its rather like a Digg.com system with a yes vote, and no redrawal, no opposing vote, no reason as to why you selected what you did and no commenting. Infact a Digg style system would be an interesting development, people voting for the best solution to the problem, it would make democracy actually a democracy, rather than this pseudo-democracy where by we vote in the guy who we hate the least and hope he does OK.

Surely the best ideas come from the little people, not from up top, when you have 1000 minds working on a problem, who is going to come up with a better solution than the 1 person at the top? We just need a government backed method to bring this scheme in to fruition.

How can imageshack survive?

I was looking at this site today about how to hack your macBook to have Apple TV on it. Interesting stuff, however it made me want to demonstrate a point to you.
See that video? 3.7MB + Flash capsule
See that image? likely: 500KB - at most: 1.5MB according to ImageShack

OK, so images automatically load and youTube videos don't. But who would go to this site and not download the video? A minuscule percentage.

Unfortunately, this is a very common occurrence on the web, just look at any Fark.com photoshop competition. So how are ImageShack able to justify this inability to work lack of bandwidth? I really don't know. But the solution?

Never use ImageShack, especially when you are expecting a lot of bandwidth.

Failing the Turing Test?

These days, human tests now common place on all popular websites, from Digg, to Hotmail, Yahoo, to SomethingAwful, and it is never so more important than in web2.0 websites where a single user can have a huge effect.
Usually, to test this code, most websites employ a technique where by their servers produce 'non computer readable' text, and look a little like this:
The premise behind this is that these letters are just about human readable - although I know it is a common occurrence for humans to get it wrong - but your typical OCR reader would have great difficulty figuring out where all the letters begin and end.
The server's job in this process is to create a code, and add each layer to generate the image. This is of course, a very intensive algorithm, and has the great disadvantage of being maliciously used to waste bandwidth and CPU time.

However, I recently stumbled across this great site Gullible.info where they amusingly post almost plausible 'facts'. However, the bit I am interested in is their feedback page which asks you to prove that you are human.

In a similar vein to Microsoft's Hello Kitty project, they use images for you to discern from a randomly generated word. Whilst this is far from fool-proof in the Gullible version, simple adaptations could make this a mean driving force with little processing power, I propose a system where the images are taken from Google Image search, based on 2 search terms that are listed and compared against.

It will allow automated, computer secure verification with low bandwidth usage and accurate responses. I'm amazed this kind of thing isn't being used more often.