Archive for Technology

Thoughts on Web Hosting… and a Plug

I’ve spent a fair amount of time over the years being a webmaster in one way or another. And in that time, I’ve dealt with a number of web-hosting scenarios, from a Solaris box in my basement (my ISP loved that one) to Co-Op type environments, to corporate webservers at various companies. I’m a huge fan of  “low cost” hosting, but I recommend avoiding the free hosting companies – spending a few bucks will save a ton of headaches later, especially if you depend on your website to generate any kind of income or business leads.  I’m sure there are exceptions to this rule of thumb, but I simply don’t have the patience to try every company and compare them.  And moving real websites you care about can be a real pain, especially if you have other applications (e.g. web log processing, mailing list services, databases, etc) integrated with your website.

There are tons of companies out there that will host one or more sites for you for well under twenty bucks a month. They’re all priced differently – yes, I know… usually in such a way as to make that one feature you need only available in the more expensive plans. Ask questions. Or if you don’t actually talk to anyone (and it’s a good idea to find out if you can talk to someone – don’t assume this!), take a good look at what the company offers. If you can go in with a list of technical requirements, all the better. You can find a ton of articles on the web about what you should want in a web host, so I’ll stick to the couple of points that I’ve found have mattered most to me (mostly the hard way):

  • Uptime Guarantees – does the company make any claims about their uptime?  Even if so, do they have a guarantee?  Does the guarantee have teeth to it?  Is the company financially accountable to you if they’re down?
  • Support Response Service Level Agreement (SLA) – how long do they take to respond to a cry for help?  Again, do they guarantee it?
  • Support Options – this one is probably most important to me.  I’m technical, so I’ve always thought, “Nah, I’ll be fine – I’ll figure it out.”  It’s their server, and no matter how technical you are, you may not be able to figure it out.  And even if you can figure it out, you probably don’t have the permissions to fix it.  You’re paying them, so make sure there is an actual phone number you can call.  And at the risk of being a little bit culturally insensitive, find out where these phones are answered – the more local, the better, in my experience.
  • Shell Access – if you’re technical, this can be a significant differentiator.  If you’re not, ignore this point entirely.  Bear in mind that if you have shell access, you’re almost certainly “jailed” in your home directory, and lots of files (particularly configuration files) will have integrity protection set up.  This can be loads of fun – you make a configuration change, and the next day, the file has magically changed itself back.  Call your host if this happens – they might be more flexible than you think, depending on what you want to change.  This is one reason I love the host I have now.

I’ve had my share of good and bad experiences.  I’m currently with HostGator* and I love them. Mail for my domains is not with them – not because they do particularly badly with mail, but because GoogleApps does a great job, offers flexible access options for users, and the price is right (hard to beat “free”).  But that’s a whole other post.

Do you have a favorite web host? Leave a comment.  And if you leave a link that gets you paid as an affiliate, please be up front and say so.

* in the interest of full disclosure, this is an affiliate link, and I get paid if you sign up through it.  Which, of course, I encourage you to do. 😉

TinEye and the New Firefox Plugin

Ever wonder where other copies of an image are on the web?

Ok, let me ask that another way: have you ever wondered where other copies of your images are on the web?

Enter TinEye – it’s like google image search, only backwards.  Or maybe “inside out” would be a better way to describe it.  TinEye roves the web making digital fingerprints of images it finds.  Then it allows you to feed it an image (either by URL or by upload) and tells you where other copies of that image are published on the web.  Even if they’ve been re-cropped, re-sized or photoshopped.  Important distinction: it doesn’t find images that are similar, at least in the human-brain-sense – an image has to be a direct digital descendant to be a match.  More info in their FAQ.

This is handy for those of you who are professional photographers and depend on people buying your images, rather than lifting them for commercial use.

My one complaint (well, ok, aside from the fact that the search engine doesn’t rate “sameness” the way your or my brain would) was that the process was a little cumbersome.  Until recently, you had to download the image (for upload to TinEye) or copy its exact URL (the image URL, not the page URL), and go to another browser window, sign in to TinEye (certain features require that you have an account), upload (or paste URL) then view results.  This only takes a few seconds, I’ll admit, but it really shouldn’t be that hard.

But things have changed – thanks to a new Firefox plugin (thanks, Lifehacker).  I gave it a try, not because I’m vain TinEye Screenshotenough to think that people are copying my images and using them all over the web (ok, I’ll admit, I was hoping I would find something, but oh, well), but because it’s just neat and I’m sure it will be useful at some point when I’m trying to find a photoshopped distortion of an iconic image.  Using this Google Images search, choosing this search result, right-clicking, and choosing “Search Image on TinEye,” I found a couple of fun permutations, some of which contained less than half of the original image.  Hint: if you want to find the most edited, distorted versions, try setting the Sort Order to “worst match.”

They also have a bookmarklet similar to TinyURL’s well-known one, that allows you to drag a URL over a button to search (only in some browsers – your mileage may vary).

Another caveat: TinEye claims they have indexed 1.215 billion images, which has to be an infinitesimally small portion of the intertubes.

Let’s see how long it takes the nice folks at Lifehacker to notice I stole their screenshot…