Browser fun and games (a little rant)

Much of what is possible with web services and the internet is dependent upon the browser that you use. This is unfortunate since there are now a large number of well understood standards so in theory it shouldn't matter which browser is used to access a web site: but some browsers, Internet Explorer in particular, are not particularly compliant. The so-called modern browsers: Firefox, Chrome, Safari and Opera, do a pretty good job however at compliance and in fact they tend to vie with each other in both conformance and performance terms.

Why does compliance matter? Well the main issue is that developers have a really hard time attempting to compensate for the quirks and idiosyncrasies of non-compliant browsers. Making a complex web system look the same no matter what browser is being used can be very challenging, time consuming and therefore expensive - and IE6/7 are the biggest problems in this respect. At Enmore Services we took a view some time ago that we would no longer support IE6 as standard, the additional costs and aggravation were no longer worth the effort.

Now understandably IE is still (just about) the most used browser, since many users are 'locked in' to it just because it was what was installed when they acquired their PC. But most users do have a choice, and while some users may not be inclined to install another browser, they could at least ensure that they have the latest IE version, currently IE8, which is far more compliant than previous versions. I find it quite amazing that some users are still 'stuck' with IE6 - why do they persist with software that is 10 years old when a much better, free alternative is easily available? Most people change their car more often than that.

How much longer IE users will have choice however is in question since there is a rumour that IE9 will only be available for Windows 7. Which seems a very odd decision, if true, since it would surely just further encourage users that are more than happy with Windows XP to start using another browser.

www.thewildernessdowntown.com

Technology keeps moving on however and the next 'big thing' in web systems is likely to be HTML5 - which is the next major revision of the HTML standard that is currently under development.

The thumbnail on the left is from an excellent HTML5 demo that I recommend you look at to see what HTML5 will be able to deliver. Click the caption to see it.

This demo only runs on Chrome at present but its well worth installing just to see this :-)

Unfortunately, already HTML5 looks like it will be the next battleground in the browser wars, with rumours and assertions already being put about. This article for example provides some interesting views on how some of the IE HTML5 pitfalls can be avoided by using XHTML5.

However things turn out with HTML5, what we can be sure of is that the browser 'fun and games' are destined to continue.

Cloud computing - what goes around comes around?

Cloud computing is one of the latest buzz phrases for a hot new technology, but is it all that new?

Back in the late 1980's I worked at GE Information Services¹ or as it was variously known then, GEIS or GEISCO. GEIS used to sell what it called teleprocessing services, which were essentially the direct descendants of the old time-sharing computer bureau services. By this time computer bureau services were more or less a thing of the past and we were obviously way beyond the famous Thomas Watson misquote² of "I think there is a world market for maybe five computers,"

GEIS however just happened to connect its two or three large computing facilities to a proprietary, private, global network, and this allowed it to become the largest global supplier of EDI services and a range of specialised services for financial markets, trading, shipping, etc. This was of course in the pre-internet world and long before the WTO deregulation of telecommunication services³, but arguably, GEIS was back then a type of cloud computing provider.

Cloud computing therefore, in my view, is just technology evolving at its usual rampant pace - combining many different components: ubiquitous, high speed network access, simplified and standardised protocols (IP etc.), standardised, low cost processing (blade servers etc.) with perhaps virtualization providing the last significant piece of the puzzle that enables what we today call cloud computing.

Now don't get me wrong, there are clearly many additional developments that enable cloud computing to easily scale, on demand, and therefore start to become a utility-like computing capability where we just do not need to know where the servers are, or how they are run - in just the same way that we don't care which power station is actually generating our electricity. But I would suggest that the main impact of cloud computing and managed hosted computing in general is a commercial one.

Just 10 years ago if you started a new software related business one of the first things you had to do was buy some expensive servers plus all their associated software, and you also had to hire some very experienced (and expensive) systems folk to configure and manage these computer systems.

There were lots of choices to be made - and wrong decisions could be expensive if you had to swap out and replace inappropriate equipment - and it generally took many months to get it all set up properly.

Today we can do all this in just a few hours - from scratch - by using hosted/cloud services that let you provision new Virtual Private Servers at will and resize them 'on the fly' to meet increased load demands. They come with a standard, but highly functional, software environment, typically a Linux operating system, Apache web server, MySQL database and PHP web development scripting tools. This standardised, so-called LAMP environment, supports 1000's of readily available applications and provides a development environment used by many software communities.

But these services are not only easier to use, they are low cost and can be acquired on a simple 'pay-as-you-go' basis without any initial capital costs nor any expensive dedicated, operational staff.

When setting up Enmore Services a few years ago this was the way that we went from the outset and we haven't looked back. So whilst there are undoubtedly some business uses that demand a dedicated environment run by your own people I suspect that  the reality today is that they are very few and far between.

1. GEIS is now GXS and in June 2002, GXS was acquired by venture capital firms Francisco Partners and Norwest Venture Partners from GE and now operates as an independent firm.
2.  Although IBM's Watson is well known for his alleged 1943 statement: "I think there is a world market for maybe five computers," there is scant evidence he actually said it. See  http://en.wikipedia.org/wiki/Thomas_J._Watson for more.
3. On February 15, 1997, sixty-nine governments signed an agreement seeking to liberalize the world telecommunications market

Opensource software - a revolution or just a side show?

I first became intrigued by the opensource movement ¹, ironically when I was helping to set up and develop a VC funded software business. It has become quite common place for opensource software tools to be used by most software developers/companies, and indeed the Internet would not function today without major tools and operating systems like Apache and Linux, but application software has remained the province of companies that ask you to pay a software license and more often than not an annual support fee as well.

The usual argument that application software companies raise is that you shouldn't trust your important software applications to a "bunch of amateurs" - but as I have progressively discovered this is far too simplistic and in many cases the argument is completely turned on its head.

Putting aside the commercial advantages of no software license ² fees, I would argue that use of a product from a well founded opensource software project is actually a lower risk than use of a traditionally licensed product from a small software company.

A large proportion of software companies have less than 100 employees, which means the development team may be very small, perhaps less than 15 people, and they always seem to have more urgent priorities than your reported bug or new feature request. Also how many businesses that start to use a proprietary software product actually bother to negotiate a full escrow arrangement that would release the software code if the company went bust - and if the worse case situation occurred what would you do with the code anyway? Contrast this with a significant opensource project which may have hundreds of developers and the source code is always freely available.

As you may have guessed I have become quite evangelical about opensource and indeed have based my company Enmore Services on the use of one specific product - Tiki CMS Groupware.

I had been using this software for a number of years (actually to support my family history research) and had become a keen, albeit a slightly frustrated user. My frustration was rooted in not fully understanding the application - but I had recognised its potential, so as part of the development of Enmore Services I became more deeply involved in the wider development community - and have indeed become a developer.

I have to confess that at first it was a bit of a culture shock for someone that had been involved with the traditional IT business for 35+ years, because I just couldn't work out how it actually worked! But it does - and is an intriguing example of collaboration that I will undoubtedly return to in future posts - and it works extremely well especially if you are prepared to contribute very actively yourself.

My view therefore is that this growing movement is certainly not a sideshow - and this view was recently reinforced when a friend recently sent me a link to a paper from a firm of lawyers where the central theme addressed is that opensource:

"has today achieved such a key role in software development that it may begin to enter a level of maturity in which competition law risks could arise".

Far fetched? Or the consequence of a real revolution?


1. See http://www.opensource.org/ for more information on opensource
2. See http://www.gnu.org/licenses/ for details on the most common GPL and LGPL software licenses and Creative Commons for the license often used for documentation etc.

What happened to the Virtual Corporation?

Back in the late 1980's and early 1990's there was a lot of discussion (1) about a new type of business called a "Virtual Corporation".

At first this was very much a "futuristic vision" but it soon became an economic necessity for most large organisations as continuing globalisation in the last two decades has driven the necessity to exploit lower cost economies. This in turn necessitated the development of sophisticated supply chain management, and in due course cooperative development with key suppliers. The evolution of these sophisticated systems fostered the basic principle of outsourcing but did the more innovative virtual company concept take off?

I'm not sure that it did in the way everyone thought it would. The core principle of the virtual company approach is that a business could be formed, that controlled the definition and quality of their product or service, directly managed the demand cycle from their customers through to a well managed supply chain, but outsourced all other activities.

Whilst outsourcing has clearly proliferated in the larger corporate world, I suspect it is in the small business sector that  the principles of the virtual company are actually being used more extensively - and this has almost certainly been enabled by the Internet and so-called "long tail economics".

So as even more sophisticated and inexpensive IT/communication capabilities become available I believe it will be the small business sector that will further exploit the powerful virtual company concept.

I'm actually making a double-bet that this will be true, by not only starting my new company Enmore Services on this basis - but also basing the company's service proposition on the delivery of web services that support the typical processes needed to operate a small/medium business 'virtually'.

I'll keep you posted on our progress!


(1) see for example "The Virtual Corporation" Davidow & Malone, HarperBusiness 1992 ISBN 0-88730-593-8