Windows 7 has been out for the last few months, and has received far better reviews than its predecessor: Windows Vista. The technical reasoning behind this is pretty simple, and well discussed. Windows 7 fixed the issues that people complained about with Vista.

Wikipedia lists the common criticisms of Vista as:

•    Hardware Requirements
•    Licensing
•    Cost
•    DRM
•    User Account Control

This isn’t really news though. People have been whining about Vista for years. Psychologically there are a couple of interesting issues though.
The cost of Vista was always going to be an issue. Most people get their OS with their computer. Having to pay anything for it is going to be an issue. If you also take into account the rampant rise in the number of computer users in the time between the release of XP and Vista (one example is families moving away from the ‘Family Computer’ model to each family member having their own machine), you have a lot of people who aren’t used to having to buy operating systems at retail price. The OEM price of Vista was pretty much the same as that of XP (I checked at the time).

This is where the social psychology part comes into play though. The only way you learn about a product is if you actually use it, or if you get told by someone you trust to know. So who exactly are the early adopters? Who’s telling the masses what they should believe? I’m sure Microsoft has a more accurate profile, but I’d imagine it’d go something along the lines of this.

The first group would be technical professionals. People who are paid to know technology, and generally don’t have to pay to get their hands on it. I’m thinking here of TechNet subscribers and people who work for companies with Volume Licensing of some kind. They have legitimate access to the software as soon as it is released, and generally play with it at the same time. There is an upside here in that these are professionals who should (in theory) know what they’re talking about. Hence their views on the product would hold water. There is also a downside though. People working in these contexts have to think of the effect an upgrade is going to have on the company as a whole. They have to make sure that all their company’s software will work (which it often wouldn’t), that workstations would be able to run it (which they couldn’t) and that the value of upgrading would mitigate the costs of re-training the workforce to deal with the changes. This forms part of what is known as Total Cost of Ownership (TCO). Based on these figures a lot of larger companies decided not to upgrade to Vista, and this was reflected in a lot of tech-talk and blogging at this level.

The second group is made up of two parts. The first are proper journalists. Honestly I have very little faith in this group, especially when it comes to technology. For example, the so-called tech expert who writes for the local news paper once wrote that a good way to backup your Gmail database was to use some applet that cached all your emails in the web browser (as opposed to say downloading them with a POP3 client). I also remember reading of journalists who commented on the new MS Office 2007 interface, confusing it with Vista itself (the two were apparently provided together to reviewers). Now confusing Windows and Office is a mistake a user could make, but for a journalist to make it, and their editor to not pick it up just screams incompetence to me.

The other half of this group consists of tech enthusiasts. These are people who actually buy the software (hence complaining about the price), but vary largely in their technical skills and background. And they are largely where Vista’s problems started.

The final group are the pirate users. The people who download the software for free from the internet, or copy it from a friend. Practically this is probably the most influential group of the bunch, because there are generally a lot more of them than legitimate users (especially in the first few months), and they tend to be the loudest.

This brings us back to the other issues listed on the Wikipedia page. Bearing in mind who the commentator’s on the operating system are gives an interesting insight into their complaints.

Hardware

The 7 year gap between XP and Vista meant that a lot of these people had older machines which they had no reason to upgrade (XP needed an 800mhz processor and 256mb RAM to run optimally in most contexts). This led to a lot of people trying to install the operating system on computers that couldn’t handle it, and then complaining that things didn’t run fast enough. This was compounded by two things: a) people were told/ believed for other reasons that Vista would make their computers run faster (which ironically holds some truth for high-end multi-core machines), and b) Microsoft released ridiculous system requirements for Windows Vista.

The system requirements also led to notebook manufacturers releasing computers with Vista, but only 512mb RAM (from my experience you need at least 1gb to run it comfortably). This compounded the belief that Vista was slow and had excessive hardware requirements.

Licensing & DRM

Licensing is a slightly more complex issue. The Wikipedia article focuses on upgrade licenses as a point of contention, but my view goes a bit further. Windows Vista lacks an edition that does not require activation (Windows XP has a Corporate Edition which does not need activation). Thus the software underworld could not pirate it with the ease they were used to, and came up with work-arounds and hacks which were less than elegant.

On a broader plane Microsoft released 6 versions of Vista (Starter, Home Basic, Home Premium, Business, Ultimate and Enterprise). These were also replicated between 32- and 64-bit versions. While this probably didn’t lead to the perceived failure of Vista, it didn’t help. Not many people know the differences between the versions, and often OEMs tend to sell the lower feature editions with cheaper machines (for example my laptop, the Gigabyte M1022M comes with Windows 7 Starter, while it can run 7 Professional perfectly with Aero enabled). This can lead to confusion and perhaps disillusionment with the operating system/ hardware. (This is something I hope to do practical research on at a later stage)

The DRM issues form an interesting conundrum. On one hand there are the legitimate complaints around fair use and consumer rights for products that have been legitimately purchased. The other hand is again that community who does not want to legitimately purchase anything. Both of these groups are by definition placed against content producers, and Microsoft stands to lose more by pissing off content producers than a relatively small proportion of their user base.

UAC

User account control was honestly a joke, but a particularly bad one. To understand the background though we have to go back to Windows 9x though. Windows 9x was basically a GUI built on top of MSDOS, and was almost by definition designed to be a single user platform. Hence it had no inherent security measures (unlike NT and *nix which were designed to be multi-user platforms). In a single user platform every user and application has full and unrestricted access to the computer, its hardware and its files (this is one reason why older operating systems fall prey to viruses so easily). In a multi-user environment the system’s integrity needs to be maintained for all users, and hence each user is given only the access they need to do their job.

The problem arose during the shift to Windows XP. XP is based on the Windows NT kernel, which means that it is a multi-user OS and held the necessary security framework. However, years of sloppy application writing and bad user habits had led to applications that needed Administrator level access to run where it should not have been required, as well as a common practice of user’s running as Administrator. This led to a lot of unnecessary security problems with malware under XP.

UAC was overtly implemented with Vista in order to prevent these security breaches. Even a user with Administrator level access has to acknowledge when an application wants to make such changes. The covert goal however, was to get programmers to start following Microsoft’s best practice guidelines.
For the end user though, it was mostly just an annoyance and quickly turned off.

My Own Experiences

I ironically fall somewhere into the early adopters group for Vista. In the first year I experienced it in depth on three different machines with differing results.

Soon after the actual release I purchased a laptop which came with Vista Biz edition (Turion single core processor and 1gb ram with an NVidia 6-series chipset). The machine was quite powerful for a laptop and did not have any problems running the OS, or the range of applications I threw at it while I was in honours. (It did have a huge problem in that it was huge (17” display) and once I got rid of it after a year simply because of the damage it was causing my back).

My brother also purchased a laptop at the same time (Sempron, 512mb ram with an ATI chipset). This machine crawled with Vista and my brother quickly downgraded it to XP. However, he later upgraded the memory in the machine (first to 1gb, then to 2gb). Once SP1 was released he put Vista back on the machine and has run it more or less happily till the present.

The third machine is that of my business partner (though I had a similar machine at the time)(Athlon 3500+, dodge motherboard with Nvidia 6-series chipset, 1gb/2gb ram). He bought Vista and had hardware issues from the word go. The service packs didn’t do too much to fix the problems, though he did stick with the OS if I remember correctly. I tried it on my machine and ran into similar problems (I stuck with XP however).

I did upgrade later on when I upgraded my machine (Athlon X2 5400+, 2gb ram, ASUS MB). I found that Vista ran better than XP did on that particular setup as well. This was however Vista with SP1.

I have since upgraded the above machine to run Windows 7, and it runs pretty much the same.

In closing though, I want to make reiterate a few points:

1. My experience of Vista was overall positive. You’re allowed to disagree and I really hope people will share some of their firsthand experiences.
2. I had the original idea to write this article over a year ago after seeing Microsoft’s Mojave experiment. Scientifically it’s about as valid as the ‘Does he love you?’ test in the tabloid, but it does provide some nice questions and starting points.
3. Any operating system has similar teething problems when its released, especially when it’s the first OS based on a particular platform (Vista was the first incarnation of NT6)
4. Windows 7 is not as magical as some would have you believe. It’s based on the same codebase as Vista and Server 2008. (Its version is NT6.1, though its build number is 7600 which I guess is where they got the name). A lot of the problems people had with Vista are still in 7, but hardware has grown up over the last three years and can handle things better. Windows 7 is admittedly lighter on resources in some areas. (I would never run Vista on my laptop for instance).

As always, your comments are welcome...

Hi all,

Thanks a lot for your comments and encouragement. I'm sorry for the lack of recent updates, but I had a crazy six months of work, followed by a few months of sickness. I will try to post something interesting by the weekend though.

In order to answer some of the questions that were raised:

1.) The site runs on BlogEngine.Net. I looked at quite a few different blog engines before settling on this one, which I felt had a user friendly interface and the right set of features.I did try a different engine before, however that crashed on me within a month, and I had to repost my original entries under this new platform once I set it up. It's using one of the default templates that comes with the installation (though I originally did intend to design my own theme). Any questions involving the engine itself should probably be posted on the BlogEngine.Net site itself.

2.) My information comes from disparate sources. The branding and identity theory is based on the work I am using within my masters. (For those interested, I am trying to create a theory of non-rational social identity based around value-systems. I am drawing on Charles Taylor's work but re-reading it from a social constructionist viewpoint)

My technical background comes from my experience with both Linux (I've used it as a desktop operating system on and off for the past 10 years, as well as a server operating system almost continuously during that time), as well as using Windows in most of it's incarnations all the way back to the day's of 3.11. I am also a certified MCSE, the study for which gave me a lot of experience with the backend of the Windows world.

3.) My primary contact means is via email. my address is mike(a)cydonia.co.za. Anyone wishing to offer more detailed commentary or ask questions is welcome to do so through this means. I will also look into implementing forums (and finally fixing the style sheet issue on the main site).

Happy reading,

Mike

Interestingly, the most expensive part of buying a computer is not necessarily the hardware, but rather the software that you run on the hardware. The hardware for a basic office machine can cost as little as R3500, while Windows Vista Business and MS Office Basic Edition will put you back by as much R4000. The second temptation here is thus to turn to open source software to bridge the gap.

But will it work for your business?

The short answer is maybe, partially, but probably not.

From the top though is the story of Linux: Basically this guy (Linus Torvalds, the creator of the Linux project) decided to create his own version of the closed-source software (UNIX) that was then run on large mainframe-style installations. What he wrote was the kernel (broadly synonymous with the engine in a car), which was then merged with the existing GNU project. The GNU project had already written and released a lot of the other core operating system components (such as system libraries, compilers and utilities) under an open source license. Practically think of it as literally an engine. It can go at 250km/h, practically though it's just pretty useless without the rest of the car.

Just to be clear though, to specifically call an operating system running a Linux kernel Linux is a bit of a problem. The term that the GNU project prefers is GNU/Linux (though I am going to stick with calling it Linux to avoid reader confusion). It does also bring up the core strength (and weakness) of the Open Source operating system.

The Linux operating system is more of a process than a product in many ways. The kernel, system libraries, user applications and everything else comes into being as community initiated projects to full a certain need that a certain person (or group of people) see. They then release the source code to the internet (and hence to the world). These individual projects are then (often slightly adapted), compiled together and packed into a distribution. This is what you then install onto your computer in the form of a CD/DVD. Ubuntu is the current golden distribution, though there are others like Red Hat, Debian, Slackware, and so on.

The upside of this process is that your end product is the work for thousands of really smart people, with a public arena to discuss aspects such as security and features. The downside is that it lacks a single guiding vision around issues interconnectedness and future development.

Windows on the other hand is controlled by one company (Microsoft), and has specific people guiding its vision. Microsoft also make a lot of the other productivity software that business uses (such as Exchange Server for email, Sharepoint Services for collaboration and Active Directory to bring almost everything together). Things are built together, and (arguably) do work better together.

So, how does this effect a small business? Here's my argument for Linux:

1.) It's free. That means that you are only paying for the hardware, which you can spend more on instead.

2.) A decent Linux power-user is immeasurably better at running things than most de facto Windows administrators in a small business environment.

3.) Everything you need for basic productivity (Office suite, email client, web browser) comes with the distro.

4.) Most other popular Windows apps have Linux equivalents.
5.) It's free...

The downside is this:

1.) You will still have occasional hardware support problems (in Windows terms not having drivers).
2.) Decent Linux power-users are hard to find.
3.) Specialised applications most likely don't have Linux equivalents or Linux versions.

Linux and Windows work very differently in their learning curve. While both take similar amounts of time to get used to the basic desktop and application set, it's pretty easy to learn to do power-user activities on Windows (such as installing device drivers or printers, installing applications or modifying basic hardware settings). To do the same effectively on a Linux system often requires knowing console level commands, recompiling the system kernel and editing text based configuration files. However, once you know this you are able to do a lot more to fix and customise the system. Windows on the other hand leaves a lot of this type of functionality to the registry hive, which is graceful in theory, but infinitely complex in its application.

The argument for Windows is more or less the inverse of Linux:

1.) You probably already own a copy of Windows (it came with your computer).
2.) It's the de facto standard, thus there is better technical support for it as well as better application support.
3.) The individual applications work better with the operating system, as well as with each other.
4.) If you have employee's they won't complain about a huge change.

The downside however:

1.) It costs a lot.
2.) New versions will almost never run decently on your older hardware.
3.) You still have to pay extra for MS Office.

I specifically left out the security angle up to now because I don't feel that the two operating systems can be compared in this way. Each application has the possibility of having a security hole. With Windows you can easily say exactly what counts as the operating system (it's everything that's on the CD). With Linux this just isn't possible as everything is just packages, from the most critical needs-to-be-there-for-anything-to-run down to the games that come bundled with KDE. There is no definitive way to draw a line that says this is operating system and this is applications.

What you instead have to do is create a working definition of OS and applications. The problem here is that whatever definition you come up with will be directly influenced by your existing loyalties. Hence, if you are invested in Linux (or are paid by Microsoft) you are going to create a security definition that just proves what you already believe.

This logic pretty much applies to anything Linux vs Microsoft related. The person writing has a vested identity stake in their product, and by convincing you to join them they are again reaffirming that identity.

(In case anyone is wondering at this point, I am very much of two minds about the both OS's, and have identity loyalties to both)

Practically though here are my criteria for Linux adoption in a small business:

1.) Will your employees be willing (and able) to make a switch? This will either mean more training for them (which itself costs money) or productivity downtime as they adjust to the new system.
2.) Can you do everything you need to do for your business through Linux? Do you need Windows-only applications to make money, or are there open source alternatives?
3.) Is it really worth the time and effort? Windows and Office are released every three years (approximately), thus licenses for both work out to about R150pm per computer.
4.) Is there someone (inside your business, or an external company) who can support your Linux-based computers? Good Linux technical support is a lot harder to find than good Windows tech support.

If the answer to all 4 questions is a resounding Yes than start by trying it out yourself. There is also a middle ground. You can try a lot of the open source software that you get under Linux on a Windows-based machine. Examples are Firefox and Thunderbird (Mozilla Foundation's web browser and email client), and Open Office (Sun's office suite). Trying these applications out yourself and using them on your employees will help a switchover.

Firefox (Web browser)

Thunderbird (Email)

Open Office.org

Branding has become a new watchword in business. Apparently if you get your businesses brand right everything else just falls into place. Personally I find this approach a bit weird, though that might be because I am reading some very different books on branding for a very different reason (primarily my Master’s degree in Psychology). Now for anyone wondering where this is going and what it has to do with IT please keep reading..
For the scope of my master’s research I am using two terms to understand branding, what we refer to as Operational Definitions. Basically these are explanations on how we will use a term in the final research submission, and they are almost always heavily informed by existing literature on the subject.

The first term I use is business identity. This is a more practical term used to talk about the way a business wants people to see it. Physically this can be used to refer to the business’ logo, its stationary, advertising, even what it’s called. On less tangible level it refers to the experience people have with the business and its products, how the business treats then and what values it expresses. In more simple terms, people are going to construct a picture of your business in their head using the marketing and experience you give them, so what image do you want to construct?

The second term is branding, which I differentiate from business identity in that it is a far more abstract relationship (in that the products themselves are less and less important) and in turn is a far more personal one. That last point is what really gives branding its power. Large companies create an image of what it means to be a consumer of their products (for example what such a customer would value and the lifestyle they would live). The consumer then gives part of themselves over to this identity, and by internalising that image they make it a part of themselves.

Tobacco and alcohol companies were among some of the first to use this approach. For those who remember when it was legal to put cigarette adverts on everything, they never advertised the sedative effects the product brought (or for alcohol the feelings of euphoria and social inhibition), but rather focused on images that portray the sort of lifestyle (such as partying and being socially accepted in a group of attractive people) that we are supposed to want. The product itself is merely a way to get that life-style.

Back to the topic of IT though, it is important to remember that the brands prominent in IT are some of the most powerful in the world. There are very few people in the economically mobile parts of the world that have not heard of Microsoft, Intel or Hewlett-Packard (HP). These brands also have some very vocal proponents, who will try to sell you products made by their brand. The fun part of this aspect of branding is that they need you to buy into the brand as that action reaffirms their brand allegiances (we can’t both be wrong can we?). It is in effect reaffirming their own identity, and hence isn’t exactly an altruistic action.

The main point I am trying to make is you need to take the brand advice of friends and relatives with a pinch of salt. Just as much be aware that IT professionals have similar brand allegiances, and these can influence what they recommend (or try to sell you).

A more realistic take on this picture is that a normal user doesn’t change their hardware very often (and by change I mean take the part out and throw it away). The costs involved in this are too restrictive and there is honestly very little practical purpose (the R4000 processor you buy today will not be that much better than the one you bought six months ago). Thus the average user will be giving you advice on two or three products from a brand that he has purchased over the same number of years, and those products are likely to be in the same price range.

A technician who builds computers might give you a slightly better idea of the quality of a particular part, but again practically s/he only uses the hardware for a few minutes (compared to the days and days you will spend working on it over the years.

My advice when it comes to choosing hardware is to buy from reputable sellers. If they neglect to give complete information on a quote about the contents of a computer chances are they are trying to hide something, especially if they leave out the manufacturer and make. At this point Google is your best friend. Simply search for the make and model and see what comes out. A good place to start is the manufacturer’s site and the product’s page. You can also add the word ‘review’ and see what other people say about the product, just make sure you read a couple of reviews before forming an opinion.

You can also ask for another quote from someone else and compare the two (or three or four) on a component by component basis. However, as a rule of thumb though it’s better to buy the products of a well-known internationally-recognised brand. You will pay slightly more upfront, but it will pay off over the life of the product.

Any comments will be greatly appreciated, as well as requests for future posts. I avoided the topic of brand allegiances around software this week, which I will cover in a future post. I realised in writing that it was too vast to add to this post.

When setting up a small office, or upgrading an existing office the most important thing to know starting off is What do I need?. This is a deceptively simple question as the way you see IT in a small biz environment (or in any business) is often unrelated to the reality of the business.

The two dominant discourses here are counter-posed to each other. The most common, especially if you are reading this post, is I cannot live without my computer!!!. Its alter-ego is easily represented by the It’s a waste of time and money!!! rant. Both are probably wrong for you as an entrepreneur though, and can cost you money one way or another.

The idea of a hyper-valued computer ideal can lead to being sold a machine that is either over-priced or is completely overpowered for the job it’s doing. This also often leads to the buyer not wanting to upgrade when the time is right as they already paid for two computers. The sad part here is that the upgrade computer you would buy 2 years later would probably be faster than the original overpowered machine.

The alternative of the under-valued computer is to buy a cheap-and-nasty, and then not to upgrade when the machine needs to be. Remember that the older a computer (particularly parts like the hard drive) the more likely they are to stop working.

One way to decide what you actually need (or rather to find which middle ground is right for you) is to list on a page how you use the computer in your business, and how you use it for your personal life. This will give you an idea of how much you should spend on the initial buy, and how much money you will lose if the computer does blow up. It also helps to know if you need the machine to run any performance intensive applications (such as video editing or large scale database apps [large scale here would be more than 10 000 entries, not a contact list of 100 names]). In this case you probably have a pretty good idea of what requirements your software needs to start with.

More importantly is to spend money where it matters. While your computer starts up in 5 seconds with a quad core processor, it is not going to give you any practical performance gains when you are using it to download email or write documents. My strong recommendation for the small business person here is to buy a decent monitor (and here bigger can be better, especially with spreadsheets) and a good keyboard and mouse. While you may wait an extra 5 seconds to start the machine, you will spend hours typing on the keyboard and look at the screen for twice that time.

Remember also that should your needs change after a year (for example your contact database goes into the 5 digits) you can most likely upgrade parts of the computer to suit those needs. Processors can be changed with minimal effort by a good IT professional and extra memory can be installed by you yourself.

This post particularly avoids any practical brand or purchase advice, which I will explain in the next post. Anyone who needs advice is free to contact me via the above link. Also, any requests for particular topics can be made that way.

Next week I will talk about the issues around branding in an IT context as well as your software choices on a new or upgraded machine.