A new method of Website repair

I ran a Website repair business for a while. It wasn’t very successful. Can you work out why?

I’ve hit upon an ingenious method of repairing broken websites, and I’m so excited about it I just have to share it. I hope you’ll be as excited about it as I am. First, the bullet points:

– It does no harm to your website.
– It works for all website problems, irrespective of the cause, or the technology used.
– It’s a panacea against hackers, coding errors, badly designed graphics, poor copy, bad search engine placement, it can even remedy flash sites.

My method is based on same principles as everyone’s favourite form of alternative medicine; the centuries old Homoeopathy. Here are the fundamental principles upon which it is based:

7/8ths of all website problems are caused by something I term “Psora”, or “Itch”. The “skin” of your website is covered with itchy spots, which have been inherited by all generations of websites from the very first HTML-based site created at CERN by Tim Berners-Lee, through all the intermediate generations of websites to the sparkly, Web 2.0 sites we have today. These Psora cause almost all website ailments.

The remaining 1/8th of website problems are caused by “Miasma”. This insidious disease is passed from website to website by your website’s visitors. If someone visits an infected website, and then visits yours, the chances are that simply by clicking on a link on your website, they will pass on the Miasma from the infected website to yours.

Like cures like. A website disease which has a particular set of symptoms can be remedied by applying small doses of a digital remedy which, in large doses, would produce similar symptoms. This is known as the Law of Similars, or in Latin: Similia similibus curentur (Let like be cured with like). Saying things in Latin is always impressive – I really must do it more often.

The more dilute the remedy, the more potent. I have come up with a system of increasing the potency of the cure to make it more effective than conventional website repair.

Ok, so how does it work? Let’s take a case study as an example.

An ecommerce site, http://www.somebrokenwebsite.com has a problem. Hackers have, using SQL Injection techniques, managed to hack in to the site, change the admin password, and delete a large number of products, changed the prices on others, and added some “humourous” (to hackers) products of their own. The site owner contacts me for help.

The first thing I do is spend a couple of hours on the phone with the client. This doesn’t actually help, but it instills confidence and makes the website owner feel that I am taking him seriously.

Hmm, actually, that’s not true. I first outline my rates and get the client to agree to send an unspecified amount of money, irrespective of whether the “repair” works or not. This, as I am sure you will appreciate, is an important first step in any homoeopathic remedy, digital or otherwise.

Second, I take a look at the website, and note its symptoms. Hmm, some products have been deleted, some added, and some have had their prices and stock levels changed. I consult the book of provings; I see that this symptom is caused by hackers, and check the recommended remedy: a 6C solution of SQL injection.

Now, some homoeopathic Website Repairers would simply go and purchase an off-the-shelf remedy, apply it to the website, and charge the client for that. I believe that the personal touch is important, so all my homoeopathic remedies are prepared individually by hand, to order.

I first construct a malicious SQL statement which is in all respects the same as the one which the hackers used to break in to the site. However, this SQL statement, if applied directly to the site unaltered, would cause even more damage, so obviously it can’t be used in its current form.

The next step is to dilute the SQL statement to render it inert; For example, if the SQL is 80 characters long, I “dilute” it with 80 * 100, i.e. 8,000, random characters. (Actually, the characters are pseudo-random, since they are generated by a computerised algorithm, but since we are dealing with pseudoscience anyway I don’t think that’s too important a distinction).

Ok, that’s a 1C – 1 in 100 – dilution. But that’s not strong enough. I need to dilute the remedy five more times. My method of increasing the potency of the cure is simply to dilute the digital potion even more. The less of the remedy in the final “cure”, the more the potency. So I take 10 characters at random from the mixture, and combine these with another 1000 random characters. I do this another 4 times, until in the end there is an infintesimal likelihood of there being any of the active SQL in the final mixture. I then take 80 characters from this final mixture. This will be the curative which I will apply to the ailing website.

Aha! I hear you say; if there is none of the original SQL in the final statement, how can it have any effect? This is where the final, magical homoeopathic analogy helps us out. The characters which have been used to dilute the bad SQL now contain none of the original, but they “remember” having been in contact with it. They retain a mystical, magical energy which is a kind of metaphysical mirror image of the original malicious SQL statement. Huh? Not scientific enough? Ok, it’s all to do with quantums. Better?

Anyway, on creating the potentised dose (the weaker it is the stronger the effect, remember), I then apply it to the website. I dip my mouse into the random characters, and then gently brush it over the website, taking care to visit as many of the corrupted products as I can.

Most websites are cured by this method in a single application, although in severe cases (such as yours) the cure may take several applications, each one being individually prepared. Ideally (for me) a cure will take a couple of dozen applications, over the course of a year or so. By the time the client realises the “cure” doesn’t work, the website is dead, his business in ruins, and he has no money left (having spent it all on homoeopathic website repair) with which to sue.

I think it’s a winner; after all it works for “real” homoeopaths. What do you think?