Posts Tagged ‘Web 2.0’

Quest for the Single Sign On

Friday, December 26th, 2008

Like most Internet users, I am plagued by the Multiple Sign On problem. I’ve got a username/password pair for GMail, Facebook, eBay, PayPal, my blog, my Coppermine gallery etc, etc. Trying to manage all these is always a bit of a handful, especially for those I don’t use often. Most sites makes it a bit easier nowadays by using an email address as your user name, but it is still a bit of a pain in the ass.

Most of the big names on the Internet has been trying to unify the logins for their services for the past while, as long as you stick to one brand, it works. For example, a Google Account covers you for GMail, Google Docs, Picasa, Blogger, Youtube and many others. Similarly, your Microsoft Live account covers you for MSDN, SkyDrive, XBox Live! etc. etc.

Howevery, you are still tied to one company for each login.

It was a pleasant surprise when I came upon Stackoverflow a while back. It is a batch of new websites using the OpenID system for logging in.

OpenID Logo

OpenID has been around for a while now. It is, unsurprisingly, an open standard for various sites to supply your login information, and thus online identity to each other, done with your consent of course. This means that once you are logged into site A, instead of logging into sites B, C and D, you can simply ask them to check with site A with regards to your identity. This is definately a move in the right direction toward the Single Sign On.

A side effect of more wide spread Single Sign On would be that it discourages internet anononimity. If people are using the same identity on multiple sites, they are more invested in that identity. This may be a good thing or a bad thing. However, I do think this will close the gap between acceptable behaviour in real life and online. That means less trolling and better YouTube comments! (Ok, I can dream).

In fact, a surprising number of sites already offer OpenID, including AOL, Microsoft, MySpace, Yahoo and most recently, Google. Chances are, you already have one OpenID… or five. In fact, this poses a problem as most of the big names are happy to supply you with an OpenID, but does not want to accept it, or be the “Relying party” as the lingo goes. This means you are still sitting with a handful (instead of a bucketful) of accounts. Furthermore, the fact that you already have so many OpenIDs is itself a problem, as there is no way to merge two IDs.

Google and Facebook have also introduced their own take on the Single Sign On problem. Google Friend Connet and Facebook Connect are services offered by two of internet’s biggest brands to users to use their logins all over the internet. In one sense, this means that Google and Facebook want to become the identity brokers for the internet.

This is one area when you don’t really want too many choices. The network effect is very real here. What we want is a handful of logins that potentially covers everything. (Note that I say potentially because you should still have the option of having unconnected accounts for privacy reasons). The situation will only get better when the big names start accepting each other’s logins. We want to arrive at a situation where your Google login is your Yahoo login.

The current state of affairs means that it’s a lot easier to post on werid forums you come acros without having to register a new account, but you still need a few “Big” accounts.

Update:

I just sorted out OpenID for Aspherical Boundaries… Looks like I was really late to the party :P.

Who writes a site from scratch?

Monday, December 22nd, 2008

Let me start by saying that I never learnt much about how websites work. I know some basic HTML and know about CSS. I have some idea about what Javascript and PHP do, but never touched any of it.

Despite this, I have built a small number of sites over the past 7-8 years, starting with a little two-pager for CS class back in high school. I started in notepad, and moved onto DreamWeaver, but still building static pages, from scratch. Every time, I would have two things on my mind:

  • I need to learn to do this properly;
  • There’s got to be an easier way…

Needless to say, I never learnt anything about CSS or HTML. More importantly I guess, I know nothing about good web design.

Why do I blabber on about my boring experiences of making websites? I suspect it is similar to many people’s site building experiences. Looking around the web, It would appear that the vast majority of sites are static sites hacked together in an afternoon. Most of these are once-off sites, and are no longer updated. However, there are those too that are continuously updated into gigantic sites. Of the photography sites I visit often, numerous ones fall within this category: Ken Rockwell, NaturfotografByThom and the monstrous Photography in Malaysia.

These hacked together sites are generally usable, though most have poor design. More importantly, they require immense dedication to maintain for the often one-man teams that runs them.

When I finally gave in to the blog, and started playing around with WordPress, I realised that there is very few situations when someone actually needs to build a site from scratch anymore. Just as most developers don’t write their own linked list class, most people making a website no longer need to be touching HTML.

Looking at the Wikipedia page on current content management systems, I suspect there are very few needs not covered by an existing solution. In fact, the last statement remains true even if you stay within the standard Linux-Apache-MySQL-PHP hosting environment.

Most of these take a few hours to set up, at most and allows the webmaster to present content in a much more professional manner than the afternoon hack job.

Why do we still see so many badly, but painstakingly put together sites?