I remember my brother asking for my help because he couldn’t get a program to install. It was called Tree and it enabled “branches” of data folders and subdirectories. It was a 25 year old MSDOS program and he was trying to use it on Windows XP not knowing what it was even for!

Another popular myth is that watching videos online doesn’t account for downloading data. Some people get really upset when they discover the opposite.

The state of the art in enhancing images is not quite as clear-cut as you make it sound in “#11 Zoom and Enhance”.

Yes, Hollywood has turned it into a joke, but statements like “it is super easy to test this at home: just open an image, and zoom … in” and “when you zoom into a photo, all you are doing is making the pixels bigger” aren’t quite the full story. Absolute statements like that annoy me almost as much as the stupid levels of zooming seen on film. That’s a shame as the main point you were making was good.

For still pictures, the comment isn’t far off: it depends on whether the image quality was limited by the number of pixels or the optical characteristics of the camera. If the latter (slightly out of focus, motion blurred, diffraction limited) then there may be things you can do but it won’t be much.

It’s where you write “they can’t extract information that is not in the file” that things get interesting – particularly if the file was a video. Video is a whole different issue. There is information in the file that’s not in a single frame. A good algorithm can merge information from multiple frames.

There’s a whole field of study called “superresolution reconstruction” which is precisely about extracting features that are smaller than one pixel in one frame of the original image.

Searching the web, I can find lots of scientific articles but not many demos. I did eventually find some from commercial sites (I’m not affiliated with any of these). So, with the caveat that commercial demos may overplay the utility of these techniques, have a look at:

Infognition’s introduction to superresolution

Infognition’s demo of different reconstruction algorithms

MotionDSP’s demo of their Ikena forensic tool

So, yes, when Hollywood zoom into the reflection of a face in a vase where the entire face was less than one pixel in the original image, they’ve lost touch with reality. However, when they enhance a vehicle’s number plate where the letters weren’t quite visible on one frame of original video (you could tell there were letters there, but not which letters), it’s not quite so unbelievable.

Nice post. Loved it!

Being the huge Matrix fan that I am, I should point out, that hacking scene is from The Matrix Reloaded. Secondly The Bourne Ultimatum also shows hacking using nmap.

@ regularfry:
Worse: Windows systems have a little trick to improve searching/opening performance of files in the various folders called “My Pictures”, “My Movies”, “Documents & Settings” etc. It’s called indexing and caching.

Unfortunately, if you have a lot of files relative to the computing performance of your machine (particularly RAM) that indexing slows you down a lot. Worse still, it can even prevent your system from booting: the system may run out of RAM before you even get to your desktop. “Too many files in My Documents and Settings” is a known issue with Windows XP for example.

Luke Maciak wrote:

Same people who had the forever-blinking 12:00 on their VCR would have no issues configuring or even re-wiring their car stereo.

Yes, exactly! I don’t understand the mentality that leads to I can change my breaks (upon which my life relies on being done correctly) but changing some things on a computer might cause some real harm. Maybe it goes back to the economists’ point that humans are horrible at judging risk.

Number of files has impact on performance…

You don’t need to hit the 95%+ disk usage mark to hit this as a real problem. Get a spindle disk, NTFS format it, write a few 100,000 small sized files, and you start to get MFT fragmentation. Add low disk capacity remaining with very large numbers of files on the disk, and you get an MFT which spread across the disk. This genuinely does cause performance issues when doing any disk accessing work.
Microsoft even mention a performance hit with 300,000 files in a folder (http://technet.microsoft.com/en-us/library/cc781134%28v=ws.10%29.aspx  ).

There is also a non-direct cause of “lots of files = bad performance” (again back in the good old days with spinning ATA disks). When you’re utilising large amounts of the physical disk you also get more chance of using a bad sector on the disk (assuming your average user has never checked for physical disk errors). This in turn can make CRC errors on file access, which can drop an old ATA down from UDMA into PIO mode which is horrific. We had this happen quite commonly in laptops at work, where the old disks were prone to physical faults due to users walking around with the laptops on. Fixing PIO mode issues became a major speed boost for our users.

Apart from the ones mentioned, this is my favourite myth about computing:

illegally downloaded mp3′s

illegal music on your hard drive

Is that like music that you downloaded after breaking into a stranger’s house and pilfering their laptop to download it with?

If you’re referring to that wishy-washy concept of ‘copyright’, you might be surprised to find that no court of law will uphold a criminal charge against someone for producing a copy of some data for their own private enjoyment on a computer that they have access to, because you simply can’t prove that there is an injured party, just as if I used another computer known as a ‘digital camera’ to make an image of a recent painting for me to look at later. The only people who have supposedly been convicted for ‘downloading illegally’ or ‘file-sharing’ had been fooled into pleading guilty to some obscure civil charge(s) using that fun old game ‘good-cop-bad-cop’.

Beyond that, even redistributing ‘copyright’ data for no commercial gain and without plagiarism, has been upheld as not only not a crime, but about as far from it as you can get, as almost everyone knows how fundamentally beneficial sharing is for a healthy society.

About point #5: http://www.example.com is not a “subdomain”, it’s a hostname! “www” is the name of a computer that serves the webpages, and ftp is the name of the ftp server, and they live on the example.com domain.

@ regularfry:

That actually could be where it came from. Then again I’ve seen users who are way to young to have ever used Win95 make that mistake. Maybe they learned it from their parents though.

@ theperfectnose:

Heh, I part-time teach at a university so I do know a thing or two about that. Thankfully I’m not working for their IT department, but I used to be the webmaster for the cs dept when I was a grad student there. Fun times. :P

ths wrote:

though you seem to publish less and less

Guilty as charged. Here is my list of excuses as to why my output is shitty lately:

– DOTA2
– Warhammer
– Steam Summer Sale games
– Lack of Sleep (due to the above)
– Random real life crap (getting teh flu, dental work, car problems, etc..)
– Laziness

Last few weeks I figured I might as well crank out one longer article per week than try to rush out 3 short ones that are like “oh, hai – this is what I typed into terminal today and I thought it was cool.” I figured that if I’m writing a post, and I suddenly get violently bored with it and try to get it over with as soon as possible so I can go do something else then I’m doing it wrong. Better put it down and come back to it later when inspiration strikes than try to power through and risk burn-out.

Oh, and the whole bitter and unhappy BOFH is more or less a comedic trope, isn’t it? I mean there is not a lot of comedy in going “my job is pretty ok, my users are actually pretty decent and my problem-children make up for being computer-illiterate by being really nice, friendly and thankful”.

That said, can I haz a job at Googles where I get to play with Pythonz or Rubies all day, and users are not allowed in the building? Cause I’d totally get on that… But it has to be local, because I don’t want to move away. :P

@ NMS:

Isn’t if funny that a car (a remarkably complex machine if you think about it) is the one piece of technology that everyone seems to understand – even people who don’t own or drive them. Back in the day, everyone and their mom used to have a VCR but no one save precious few knew how to set the clock on these damn things. Same people who had the forever-blinking 12:00 on their VCR would have no issues configuring or even re-wiring their car stereo. Go figure.

As for Google – yes, that’s exactly it. But that’s exactly my point: the ability to identify good results from bad is not a “computer guy” skill. It is critical thinking skill. Sure, the sheer volume of information can be overwhelming, and a lot of links will lead you to useless places but after spending a few hours with Google you should start to see patterns like:

– specific querries with carefully chosen keywords yield better results than generic one-word queries
– seo-spam can be often spotted at a glance just from the contents of link/excerpt
– certain websites are more helpful than others (sales sites vs forums / communities)
– etc..