From my diary

A week ago I accidentally installed what I believe to be malware on my main PC, a Sony Laptop.  I uninstalled it at once, and scanned for malware using Kaspersky, but the damage was done.  Kaspersky could find no indication of malware.  But there is no reason why anyone would camouflage an installer as a PDF inside a self-extracting zip, other than to install something malicious.  So I must presume that the machine is tainted.  I use this machine for my online banking, so I don’t dare connect it to the web.  Since I don’t know what has been done to it, I can’t trust it.   It’s cheaper to buy a new machine, than to risk identity theft.  And I had noticed that the disks on the old machine tended to squeal a bit after the machine got hot, which I didn’t like.

So last week I purchased a new Samsung RF711 laptop.  Yesterday I unpacked this and began the process of setting it up.

I also bought a 256 Gb Samsung solid-state drive.  These are much faster than hard disks, and, if you use one of these as your primary drive, and put Windows on it, Windows loads very much faster, as this page informed me.  The Samsung came with an empty hard drive bay and a fitting kit.  Time was, when fiddling with hard disks could destroy your laptop.  But clearly Samsung expect you to.

Here’s what I did.

  1. I started the machine as per instructions, and ran through the Windows 7 start up.
  2. I shut down the machine, removed the battery and power cable.
  3. I took the SSD out of its package (it comes with instructions), and opened the fitting kit from the laptop.  The latter consisted of a bracket, a cable, and some screws.  There were two sizes of screw: 4 short ones, to hold the drive into the bracket; and 4 long ones, to hold the bracket to the laptop.
  4. I screwed the drive into the bracket.  Then I fitted the cable onto the drive (it can only go in one way) and the other end onto the laptop (which could go two ways, so be a bit careful – the correct way is the same way up as the other drive).
  5. Then I screwed the drive into position, thankfully without losing any screws.
  6. I refitted the battery and power, and fired it up.  The PC started normally, but I couldn’t see the new drive in Windows explorer.
  7. Then I bought and downloaded a copy of Paragon Migrate OS to SSD 2.0 ($19.95), installed it and ran it.  This was very simple, could see the new drive, and just copied the C: drive to it (including the recovery partition, I later discovered).
  8. I then shut the machine down completely, and restarted.  The Samsung comes up with a logo and a menu at the bottom, F2 for BIOS, F4 for recovery.  In the Bios I changed the boot order, so that it booted from my new drive.
  9. Save, exit, restart and … Windows 7 started, and was completely booted in 9 seconds.  Wow!

One problem that I have found is that the drive letters get a bit messed up.  However you can correct this, I believe (haven’t done it yet).

At the moment I am engaged in copying all my backed up files to a new external hard drive (I always have two), and getting the machine set up the way that I want it.  Tedious, but inevitable.

The backup is about a week out of date.  So I will need to look at the tainted machine, work out what I did, and repeat that on the new machine.

The Samsung did not come with oceans of rubbish pre-installed, which was welcome.  It’s very generic; not nearly as nice as the Sony was; but the internal 1Tb drive will be very welcome!

UPDATE: It all went rather horribly wrong later, tho — see subsequent posts.

Share

Some primers on online security – because the criminals are really there

I’ve found a number of articles online which seem to contain useful security advice.

All tedious stuff that I’d better read carefully.

 

Share

From my diary

I’ve managed to install some malware.  Oh bother.

It’s something any of us might do.  I was accessing a site which was offering the download of a book.  The link took me to a site called “blitzdownload.com”, which downloaded an exe called “<book title>.exe”.  Naturally I presumed that the thing was an executable zip. 

So I double-clicked on it, and up popped a message asking if it was OK to allow the programme to modify my computer.   Of course a zip wouldn’t do this; but I was tired, and I see that message far too often.  So I said yes.

And gradually I realised that the thing was trying to install this, and that, and the other.  I cancelled out.  But the Windows Start Menu still showed something — some sort of downloader — had been installed.

So I deinstalled it.  But of course, who knows what else it might have done. 

I started Kaspersky and checked for boot-sector viruses, and it showed nothing.  But … if it’s really a trojan — and it clearly is bogus in some way, because why else would it hide its nature? — then it may have interfered with Kaspersky.

OK, well let’s just go back to the previous system restore point.  I fire this up, make the request and … after lots of churning, it gives an error.  I try the previous restore point — same problem.

But when I restart I find that the rollback has at least partially taken effect.  The main effect is that my Kaspersky now won’t start properly.  Nor will it fix itself.  When I try a reinstall it warns I may have a virus. 

And so on it goes.  I will spare you a blow-by-blow account.  Yet here we are, hours and hours lost, struggling, while tired, just to make sure my PC isn’t about to email my bank details to fraudsters.  I don’t dare NOT do so… and yet the recovery process is so very, very fraught.

No wonder so many PC’s are infected.  Trying to make sure that yours is not, after something commonplace like this, exhausts the soul.

Wish me luck!

Share

Developing web pages in PHP

Most web space these days comes with the PHP language, usually running on the Apache webserver, with the MySql database.  When developing scripts of one’s own, ideally one replicates this on the PC.  But frankly, installing all this stuff is a faff.

This week I came across the Wampserver package.  This allows you to install all three items in one go, and has a single menu to start (and stop) the lot.  It doesn’t install a load of things that you then have to manually disable, and it just works out of the box fine.

I’ve used Notepad++ for most such scripting, but this is rather underpowered.  Instead I have been using lately Eclipse for PHP, the PDP development tools (PDT) version.  This likewise works well.  If you are a Java developer, with an Eclipse setup, you can relax — it unzips to a different directory, and doesn’t interfere at all.  You can run, thus, two different versions of Eclipse quite happily.

For unit testing I’ve been using SimpleTest.  You get the software and unzip to a directory on your C: drive.  There is an obsolete eclipse plugin for this — which no longer works and ought to be updated.  But you can run SimpleTest just fine in eclipse anyway, using the following instructions:

  1. download and install SimpleTest.
  2. put a require_once('autorun.php'); at the top of the test file.

    note: this requires the SimpleTest directory containing autorun.php to be in your include_path. alternatively, you can include autorun.php by full path, like require_once('C:/full/path/to/your/Simpletest/autorun.php');. it might even be possible not having to change the test file by including autorun.php via auto_prepend_file.

  3. run the test by right clicking on the test file and select “Run As PHP Script”
  4. the output from the testing shows up in the eclipse console

Although in my case it doesn’t show up in the console, but as a web page, so long as Wamp has been started.

Share

From my diary

Microsoft do make some rubbish software, don’t they!

This evening I sat down to work with the new Mithras pages that I am working on.  This involves running a little web server on my PC, so that I can run some PHP scripts.

Last weekend I installed (with some difficulty) IIS, FastCGI and PHP on my Windows7 box.  This evening I fired it up again … and it didn’t work.  Permissions problems, you see.

You do not have permission to view this directory or page because of the access control list (ACL) configuration or encryption settings for this resource on the Web server

What?

Now this is me, on my personal PC.  I trust me.  I am happy with anything I want to do.  I am administrator, even.  So … permissions problems?  And why now?  Why is this a new problem?

The error message gives me no idea what the problem is, and less what to do about it.  I open the management console for IIS, and am presented with a baffling array of options.

After half an hour and much Google searching, I stumble across this, which gives the answer.  I change the permissions on the home directory, add a user IUSR (why doesn’t IIS do this?) and another Everyone (ditto) and give both full control, and everything suddenly works.  But I am now too frazzled to do what I originally had in mind.

A web server is the simplest of objects.  All it does is listen on a port, and, when you ask for a .htm file, send it down the pipe.  That’s it.  That is the function that it performs 90% of the time. 

All the rest is minority activity.  We call these “edge cases”, in computing speak.  90% of the user cases is “copy file to port”.  The rest is used much less.

Yet Microsoft have produced a piece of software that fought for half an hour with a highly technical user who simply want to serve up pages to himself on his own PC.  WHY??? 

Of course the reason is obvious.  The developers have added functionality, and added it, and added it and added it until the thing has become a monster.  You could probably connect my PC to the web and run a full-fledged server on it now.  But … I don’t want or need any of that.  I just want to display some html and run some PHP scripts.  Like most people do, most of the time, who start up a web server on their own PC.

Microsoft are very prone to this problem; to lose the main user case in catering to edge cases. 

Idiots.

Share

From my diary

I’ve spent some time messing around with perl yesterday and today.  I’ve been trying to find a way that I can edit a page using FrontPage — my usual HTML editor — and do footnotes inline, using <ref>…</ref> tags, just as if I were in Wikipedia.  A perl script then reads the file, strips out the references, inserts a [1], [2], and dumps all the footnotes at the end. 

It’s sort of working.  It’s not that complex a script, although if I didn’t hate perl so much it would probably be done by now.  The language always fights you, unless you immerse yourself in it, which of course occasional users like myself never do.

It still needs a bit of work, tho.

Share

Converting a page photograph into black and white

The typescript of Ibn Abi Usaibia reached  me in the form of digital photos of the pages.  These were evidently taken under fluorescent light, since the images are huge, green, and weirdly coloured.  They’re so large, in fact, that they are hard to manipulate.

But I needed something a bit more normal.  So I was tinkering with the image and, quite by chance, got what I needed.  This tip may be of general use, where we have black text on white paper in funny-coloured photos, so I add it here.

  • Export the selected page from FR10 as an image.  Mine was a png, and came out as 32mb in size!  Here’s a snippet.  Note particularly the “see through” text to the left of the diagram.

  • Open with Paint.net 3.5.10.  Trim to right size.
  • Adjustments | Black and White, to convert to greyscale.

  • Adjustments | Brightness contrast, to turn the background white.  So increase the brightness as far as you can without losing text.   The idea is to lose as much of the background as possible, and in particular any see-through text.  The text will be very grey.

  • Then you can also increase the contrast if you like.  Juggling the two should give a pale image.  Mine were brightness=100, contrast=20.
  • Then do Adjustments | Auto-level.  This will turn the pale grey text black again.  (If you didn’t get rid of enough background artefacts,  these will promptly appear as smudges, so you may have to go back a stage, and increase the contrast – that’s what disposes of a lot of them.)  The larger the image, the better the result when converted — this image is a little small, and the text ends up a bit fuzzy.

  • You can then do minor cleanup manually of dots etc.

As someone who is quite useless at image manipulation, I thought I would pass this on.

Ideally one would save the end product as black and white, but I haven’t worked out how to do that.

UPDATE: For some reason you can’t do it in Paint.Net.  But you can in the Windows accessory Paint, which comes free with Windows7.  Just do File | Properties, change the image to black and white and save.  The file size drops from 45k to 12k.  Here’s the sample:

Note that true 1-bit black and white doesn’t resize well — hence the jaggedness in the thumbnail above — but the full size version is fine.

Share

From my diary

Last night I completed the arduous task of manually correcting all the OCR’d pages of Ibn Abi Usaibia.  Not that it is perfect even now — optically correcting is an error-prone business.

Today I moved on to the next step — getting the text out of Abbyy Finereader 10, and into some format that can be edited for layout, etc.  This is proving rather trickier than it should.

To do the OCR, I divided the 1,000+ pages up into 27 projects, each of about 40 pages.  Since the manuscript is typescript, there is really no text formatting to retain — no italics, bold, etc — so simply exporting it as plain text in HTML format, using the Windows 1252 encoding, would seem to be the right choice.

Unfortunately projects 2 and 3 are refusing to do the export.  Attempts to do so bring up programme errors, complete with .cpp file and line number.  This sort of unreliability arrived with Finereader 10, and it is an unmitigated pain.  I can’t export as Word either.  Nor can I import the projects into Finereader 11 (a truly duff version, if ever I saw one, which will rarely import any project from a preceding version successfully).

I’ve managed to export the text as unicode text format, in a .txt file.  But naturally I am rather annoyed.  The projects show no special sign of corruption, although Finereader projects can become corrupt, mysteriously.

This is infuriating, and it undermines the point of using the software.  Investing weeks of work in editing something, only to find that you can’t get your work out very easily, is quite annoying.

Finereader 8 was rock-solid.  Finereader 9 had better recognition, but was less reliable.  And so it has gone on.

Abbyy need to invest some time in improving reliability, or they will lose their market.  People who use OCR software work hard.  They should be able to rely on the software not to crash.

UPDATE: I have now installed Microsoft FrontPage 2002.  I usually use FrontPage 2000 for general editing — it is curious how neither DreamWeaver nor ExpressionWeb has a decent WYSIWYG editor, almost 10 years on — but this can’t handle unicode characters.  FP2002 can; but for some reason you cannot run both on the same machine.  And, sure enough, FP2002 has silently deinstalled FP2000, drat it.

Fortunately FP2002 has created new .htm files for projects 02 and 03, by the simple process of pasting the unicode .txt files into them.

What I shall need to do now is think up a way to format 1000 pages of text in a satisfactory way.  Particularly now that FP2002 has uninstalled all my macros!

Share

Problems upgrading from IE8 to IE9 on a Sony Vaio on Windows 7

For some months I have been using Internet Explorer 8, when I really wanted to upgrade to IE 9.  But I just couldn’t!  When I tried, the installation seemed to work; then my PC would reboot; and then it would come up in black and white, attempting to set “personal settings”, and would just hang.

Nothing I could do would fix this.  Even if I went to the Advanced Options in IE, and tried to Reset to factory settings, this would hang too.

This evening I finally managed it.  Since I never actually found a page which explained the problem, here are my thoughts.

Firstly, I usually used the 32-bit version of IE8, rather than the 64 bit version.  This was on Windows 7 64-bit.  The reason for this was that IE8 64-bit did not support flash, so a lot of sites did not display their content.  The Flash driver for 64 bit IE8 never did arrive, as far as I can tell.  But I don’t think doing the Reset on the 32-bit does any good.  You need to do whatever you do on the “main” version.

Secondly … you can’t reset IE8 while it is running.  This was the breakthrough.  Instead, close down all your IE instances, and then do Start … Run … inetcpl.cpl.  This is the Control Panel options for IE, and you can access it without IE running.  I was able to do a reset to factory settings OK!  Then, when I started IE8 — using  the 64-bit one, you may be sure — it started normally.

Then, from the 64-bit IE8, I clicked on the IE9 icon and did the install.  I closed IE8 as it ran — no reason to have it open.  I also paused Kaspersky Internet Security — which notoriously interferes with a lot of things.

And … bliss!  IE9 installed just fine.

At least Google should now stop nagging me about using an old version.

Share