Moving on to the next point

Since Google (bless ‘em) decided to stop supporting FTP for this blog, it appears that I no longer have access to an archive of the contents.  That’s only slightly disturbing, because I very much doubt that anyone will ever want to refer back to older entries.  In fact, unless it’s pushed directly in front of anyone (Facebook, I knew you were useful for something!), no-one probably even looks at the up-to-date stuff.

The point of all this rambling is that for one reason or another I’ve decided to look again at my domains and hosting arrangements.  1and1 and I are likely to have a parting of the ways in the very near future, mainly because I don’t feel that they are very answerable to their customers.  The straw that broke the camel’s back was their unwillingness to provide me with proper access logs, including full IP addresses.  All I get is the /24 subnet.  This is not acceptable – and most certainly wouldn’t be OK if it were a commercial site.  Therefore, I’ve decided to vote with my feet, or at least with my credit card.

The requirements are for a domain registration and hosting service that allows me to manipulate DNS as I see fit (not under the strict 1and1 control) and which just provides the service and gets out of the way in terms of web sites.  Oh, and since the credit card comes into it, it also needs to be cheap. 

Of the possible candidates, none stand head-and-shoulders above the rest, but I do have a certain regard for kNET: they seem to offer reasonable storage, good bandwidth allowance, multiple domains etc all for no more than 1and1 charge.  Ok, bandwidth isn’t unlimited, as 1and1 claim to provide, but does that really matter?  After all, there’s little enough traffic nowadays, given that the blog no longer resides on my storage.

Move along, nothing to see

With reference to the last entry, it now seems to be definite that the BSODs linked with MS10-015 were linked to a root kit, and are not the result of sloppy programming by Microsoft.  Rather the reverse – it was those lazy hackers that did it.  However, you’ll all be pleased to know that they have got their act together and have updated the root kit to prevent this happening in future.  Kudos to the malware authors.

Cleanliness is next to …

Given the problems with the latest batch of Microsoft updates, and KB977165 in particular, it’s been interesting following the development of comments across the Internet.

Confusion over which update was causing the issues was rampant at first, with suggestions that all the recent ones needed to be removed before the BSOD was avoided, was quickly resolved as the ‘guilty’ party was identified.  But there still remained the question about what was actually causing the blue screen.

An early idea was that the only systems suffering from blue screens were already infected with something, and it now appears possible that at least one underlying cause has been found.

However, my interest is in some of the associated comments regarding how this was missed pre-release – and this poses an important question.  Presumably Microsoft keeps there system as malware-free as they can.  Patches are tested in a known environment, and while they can be expected to take into account as many variables in terms of configuration and hardware as possible, is it fair to suggest that they should take into account all the possible malware infections as well?

Those that shout loudest are likely to be people with infected PCs, but surely it’s their own responsibility to keep the system clean, not Microsoft’s to test against an already corrupt setup.  In essence, my argument is that KB977165 is not the root of the problem; as usual, this is clearly the fault of IT Security in ensuring that no infections got into MS.  A case of “damned if you do, damned if you don’t”.

A new hope

Didn’t someone use that title elsewhere?

Welcome to the brave new world, which seems to be functional including the archives.  So perhaps Google did overcomplicate matters in their instructions, or I’ve just been lucky.  First time if so.

All I needed to do, in the final analysis, was to set up a CNAME for blog.pemur.com, pointing at something going by the slightly worrying name of ghs.google.com, and then switch my Blogger settings to a Custom Domain, pointing at http://blog.pemur.com, and it worked – after a while.  There’s even a valid RSS feed, on http://blog.pemur.com/atom.xml, should anyone find this worthwhile subscribing to.

The negatives include the loss of my favicon: I’m going to miss that.  A boring orange B that looks like nearly every other blog on the web isn’t a suitable replacement.  Also, the RSS feed that Google provides (default) isn’t exactly inspiring, either.

However, if nothing else, moving the blog should kill the overheads on my web stats for the pemur.com/.co.uk domain.  I never did have much luck extracting out the real data from the occasional spider and regular FeedDemon lookups.  On the downside, I guess it will be even more disheartening once these hits are gone.

This may be the last update for a while

Not that there have been any at all, recently, if I’m honest.

No, in common with many others today, I’ve received an email from Google informing me that blogger.com blogs will no longer support FTP updates to my own web space, and I’m going to have to transfer to a custom domain, hosted on their system – if I understand it correctly.  Apparently, FTP is too complicated for them to deal with.

Google, I’ve got news for you.  Your Custom Domain looks too complicated for me to deal with.  They promise some sort of migration tool, but when I have the vagaries of 1&1’s DNS to cope with (including the all too common lack of control) this may not work.

What’s so difficult about FTP?  You give it a target, a username and password, and tell it to transfer the files.  The data remains under my control thankfully, so even if Google can’t transfer stuff I still have it for a manual upload later if necessary.  What would future generations do without access to these gems of erudite wisdom?  (That is a rhetorical question – I have a pretty good idea of the answer…)

So perhaps it’s time to rethink the pemur web presence – I own the critical domains (.com, .org, .me.uk and .co.uk) and so far they all do different things (apart from pemur.me.uk which does absolutely nothing at all).  The original idea was that .org would be sort of work-related, and .com would be my personal domain, including biking stuff.  Now I’m back on two wheels that may well take off again, and if the Google migration doesn’t work out then I can look in more detail at managing the structures myself.

Don’t think of it as the end of an era, just the start of a new phase in the Internet’s development.