Companies always have to choose between various options to reduce costs and maximise benefits. They must decide whether to rent or buy offices, to externalise or otherwise certain business functions: with outsourcing, offshoring and other strategies that, in the end, help develop the core competence of the company without submitting it to increases or decreases in the size of the workforce or property required according to demand.

In the IT field, such decisions are normally related to software development: the typical ‘do we do it in-house or externally?’ question. But it is not very common to apply similar strategies to other areas in the IT field.

In the area of IT infrastructure (where the company’s services run), such thoughts are often dismissed because the criticality of the applications leads to thoughts that it is not a good idea to delegate their operation, not even a small part of them, to a third party for fear of losing ‘total control’. Right now, and with increasing frequency – above all in the internet company sector – there is an increasing perception that it is not necessary for the company to develop its own infrastructure to run its applications. Some even decide that they don’t even need their own infrastructure to develop their core business.

If we translate the problem to a more common equivalent – when we flick a switch to turn on a light, we want illumination and we don’t really care where the electricity is produced, we simply want the thing to work. We don’t need a nuclear power station at home to light a bulb, but if we have a company and electricity is the key to developing our business, we will surely want to maintain the electricity supply in case of failure. In such cases it would be more practical to think of having multiple failover providers other than becoming our own provider.

Utility computing: IT like water or electricity

For years this thought has been shaped by the leading computer companies who want IT converted into just another commodity, such as electricity, gas or water, for client companies.

Currently certain components such as computation or storage are being commoditised to a sufficiently mature level that they can be used by companies in production environments, something which is important because it permits economies of scale in supply companies who can therefore offer a better service at a better price.

So, we could have our application running on virtual servers which we pay for according to usage, our backups or content to serve could be on remote storage servers, meaning that we would avoid spending on backup and storage devices (the cost of a good backup system can be significant). We could serve certain content by employing content delivery networks (giving the impression of having a data centre near the client) and bit by bit externalise certain infrastructure services to specialist providers reducing costs and improving the quality of service.
We could go even further and adapt certain components to be available such that they are provided by specialist 3rd party companies. This way we would avoid the development and maintenance costs, which in certain cases can be an important part of the budget for an application. We could optimise and externalise if it is not adding value to our application.

Digital signature on demand

Within this type of components ripe for being provided by specialised 3rd parties, we find public key and digital signature infrastructure, more and more vital in applications, above all in the internet world, but which are complex and are non-trivial to develop.

In this field, there is a tendency to use existing libraries in the product or install already developed products – open source or commercial – from third parties. The problem is that development, installation and maintenance of these types of solution is not very accessible.

They’re only within reach of companies that have huge budgets and, even in these cases, they often lack the know-how or don’t budget for maintenance and continuous improvement of the infrastructure. The result is deployments that don’t represent the total cost of ownership (TCO), quickly become obsolete, are deployed in a buggy state or even aren’t deployed at all.

The field of public key infrastructure was the first of the two to make a move and now today certain activities, such as the role of issuing certificates, is limited to a small group of specialist providers, given that it is an activity that requires an large amount of money and strength to put into production.

People are still waking up to the complexity of digital signature infrastructures and while very few consider developing their own certification agency, it is common to attempt to develop a digital signature authority (be it for creation, validation, signing or storage).

A good idea is to apply a similar criteria to that of certification agencies and delegate the digital signature services to a third party and therefore concentrate on developing our main business activity.

In fact, more and more public institutions and private companies in Spain are making use of third party platforms that offer public key and digital signature infrastructure services, opening a new IT market and reinforcing the idea that such critical jobs should be centralised to specialised providers who can take advantage of economies of scale to optimise cost and maximise the quality of the end product.

We believe that this is an important idea since the domains of public key infrastructure and digital signature should be something that companies can integrate in their applications without the need for large sums of money or effort as has been the case till now. We realise how complicated it can be to solve these problems because we too have been there and for this reason we decided to open our digital signature services and offer them as third-party services because in the Internet, security should be a commodity, not a luxury.

A while ago, we described our decision to use Ruby on Rails to develop Tractis and stated that we were happy to have chosen rails instead of Java. We also mentioned that, given the deficiencies of Ruby, we opted to employ Java Web Services for the digital signature back end.
Back then we were very happy with our choice of rails and we still are. The framework has given us a lot of agility and it’s fun to work with. Anyone who follows the blog has seen that we even participate whenever possible at Rails events to show our satisfaction.

So, what of the development we’ve done in Java?

Re-reading that post, it sounds like we’re developing in Java because there was no other choice – that if we’d had the necessary Ruby libraries we wouldn’t have any Java code and we’d be even happier. With the appearance of JRuby there are those who ask us if we’d not rather run the application in JRuby on Rails and therefore have access to the Java libraries that we need, eliminating our need for web services.

Without doubt, that decision was not just about frameworks but was also an application design choice.

The design choice: web services architecture

As we said, when we selected the framework, we also chose the architecture. We decided that we wanted to develop various specific applications other than one that encompassed everything, in the pure UNIX style.

The truth is that, looking back, we’re very happy to have opted for a Rails front end but we’re also very happy to have chosen a web services architecture. So, just like when we discussed the advantages of choosing Rails, now we want to share with you the advantages of using web services.

Another buzzword

This type of architecture is fashionable and even has it’s own buzzword: SOA (Service-Oriented Architecture). So you can now have your SOA application, done in Rails with a lot of AJAX.

If, on the contrary, you are like us and you are not motivated by buzzwords, read on

Quality

As we said earlier, the idea to design smaller and more specific applications that make up a larger application follows the UNIX philosophy.

If we stop and think about it, this is the philosophy that makes UNIX so powerful and one of the reasons why we like it so much. You have a load of specific programs: ls, grep, sed, awk… these programs do specific things but they do them very well.

Working with specific applications that have to resolve a concrete problem, we can concentrate much better on the problem at hand. You might think that to achieve task isolation it would be enough to develop libraries instead of applications – and you would be right. Having said that, developing distinct applications connected by web services allows us to have a totally language, framework and deployment-agnostic architecture. If, in the future, there are better tools in C++ or .NET for the validation of signatures, we can migrate just this application without impacting the rest of the system. Developing specific applications we can employ the best tools in each case.

In addition to the improved implementation quality, we have also discovered improvements at execution time – on one hand for scalability reasons which we’ll discuss in the next section, and on the other the flexibility to redeploy or deactivate a specific service without having to shut down the rest of the services.

Scalability

When we talk about scalability, we refer to scalability of technical resources as well as human resources. On the technical side, it’s very simple. At the start, you can have all the services on the same machine and if one becomes a bottleneck, you just move it to a dedicated machine. Additionally, if the service is share-nothing, you can deploy it on as many machines as you need to give a high level of service. This will be clear to anyone who works with Rails

On the human side, a service-based architecture allows us more room to scale our development, meaning we can move from managing a single larger team to assigning a smaller team per service. This way, as we grow and require more developers, it won’t be necessary to have a bigger and bigger team but rather a group of small, agile teams that are in charge of one or more services.

In future posts

In future posts we hope to talk in more detail about the our architecture, the services that compose it and what they are used for.
Bye for now!

- Author: Mauro Entrialgo, via Sant Feliu Online.

People ask me, “Well, gee, if IE7 is starting to catch up to Firefox, and if they’ve got their hand back in development right now, and eventually they might actually catch up to Firefox in terms of features, what’s the benefit of using Firefox? Why are you guys still around if you say that your only goal is just to make the Web a better place?”

My answer to that is, how much can you really trust a company that five years ago completely left you abandoned? If they do, in fact, succeed in taking back some of the market share that Firefox has gotten back from them, who’s to say that they’re not going to disappear again? My issue is not so much at a product level; it’s at a company level. How do you trust a company that left everyone out in the cold for five years?

That, my friends, is a very important point. So important that it is worrying. That is why we use the best for Tractis: free software. We don’t depend on ASP, .NET or any corporate economic or functional ties. Imagine we use Microsoft’s databases. To feel completely at home, we continue to use Microsoft material. Why not use .NET and MSHTML and why not other things? In two years’ time, Microsoft decides to eliminate .NET or, even better, not update .Net or any other development platform simply because they’ve launched something new. Will Microsoft give away all that software so that non-Microsoft people can continue with it? The answer is NO. In this case, we’d be in dire straits:

1. We’ll have to spend more money on new versions of development software.
2. We’ll have to update servers so that the new software can run decently.
3. We’ll have to change and rewrite many things, bugs will come back.
4. We’ll have to correct the new development’s bugs.
5. We’ll have to save money for Microsoft’s next move.

No. I don’t think anyone is interested in using software that is not updated because of a company. I don’t think that it’s a good thing for a company to sit around waiting for a bug to be solved when a new function has been added. In the same way, I don’t want a browser that, when it’s launched like that, will remain in the same place for another five years.

And Firefox is not like a GNU/Linux distribution, it’s not a green product and it does not lack anything. It’s a well-finished product and considerably superior to Internet Explorer in terms of features and support. The same goes for other browsers: Safari, Opera, etc. They are at least as good as Microsoft and they have nothing to fear because they’re not competing.

It’s a pity because the version of our software rewards all the people who use browsers that support the standards. Those who continue with IE will have to wait until Microsoft gets up to date with that task. We can’t change the source code of Internet Explorer to improve your user experience.

Now is when I ask myself whether the Microsoft people actually listen to me. Not me specifically but us, the users. I read in Enrique’s blog about the release of Internet Explorer 7 Beta 3, where supposedly Microsoft has not fallen asleep. Enrique refers to Microsoft’s hassle in launching a new competitive product but, in reality, I think that it’s more of the same shit in a different package, with a new label and a new organic component that will enable us to live for five more years. Anyway, for those who don’t know, yes, Microsoft is launching a new browser. Not much is new. I’ll explain what I hate about this move.

When I entered the blessed browser site today, after reading Enrique’s post, the only thing I read was: we heard you, you wanted it easier and more secure. This message seems like a big punch in the face. The reason is that Internet Explorer 6 is easy. The new version 7 is not easy. Lots of things have been added which I’m not actually convinced as a user. Then came the never-ending text on security. This is like living in Iceland and trying to sell me ice every day. No, I’m not tolerating this. Not even a little.

On the other hand, you get the comparisons that always end up being published: Firefox vs. Internet Explorer, Opera vs. Firefox, Opera vs. Internet Explorer. You know what? What a waste of time. It’s not a waste of time making comparisons; it’s a waste of time comparing a minivan with a Ferrari. It’s a waste of time comparing a cat with a panther. Internet Explorer is simply a mirage for those who don’t understand, for those who only use the buttons, like my mother, your mother, your cousin, your friends. Deep down, you’re always using a 50s car with a 90s bodywork. I’m not sure if I’ve explained myself: you’re being sold the premise of more security, more ease of use and more functionalities but, in the end, it’s still the same 50s car. And that message was for those don’t really understand that subject. It’s no use comparing browsers by use. Their insides should be compared. ALL browsers are extremely easy to use. I haven’t seen many people lost in the steering box, return, go forward and reload. Their goals are all the same, they all copy each other trying to distinguish themselves but, in the end, their insides, i.e. the heart of the matter, are all different. And the best ones don’t exactly come from Redmont.

Then I read that many people are now experts in memory fluctuation. As if it were scarce. Does anyone actually test Firefox? This is the status of my Firefox when it opens:

5826 minid     16   0  100m  34m  18m S  0.0  3.4   0:03.39 firefox-bin

And when I have 25 windows open in different sites, weighed by the amount of Flash files, images and content, it is as follows:

5826 minid     15   0  217m 101m  32m S  2.7 10.0   1:20.99 firefox-bin

Then, what are they talking about? Does Firefox eat a lot of memory? Of course, it’s normal. Internet Explorer is not far behind either. Do the test, open the same 25 windows and measure how much they all weigh. It’s not a question of how much memory each program eats, but the amount of material we are loading per hour. If we’ve opened 25 sites, with all their images, content, flash files and nonsense, it’s normal that a disproportionate amount of memory is being used. I’m not an expert but I think that Firefox decompresses more memory than what it needs, just like Internet Explorer and other browsers.

Memory is not directly linked to the program but to the content we’ve loaded and executed at that time. Therefore, judging a browser because of its memory consumption is a bit ridiculous, it’s like opening 3,000 documents with UltraEdit or any text editor and then opening 3 letters with Word and say that it is better because it occupies 300 megabytes less than a simple text editor.

Finally, the most important thing is not the browser’s memory, its new colors or its new add-ons, but its engine. Here is where Microsoft is not fully abreast, it doesn’t get its skates on, and it doesn’t update anything. Neither in its old or current versions (IE6) or in its new or future versions (IE7). There are at least 100 features that should be changed in order to pass any minimum tests of CSS, XHTML and other things. Moreover, you should consider that we, as developers, have to do all sorts of rigmaroles and strange things to ensure that our product works on that browser. This means that the product is not 100% genuine, healthy and well done. Likewise, 100% of web sites that support Internet Explorer probably have that quality level: a rubbish code, hacks, repetitive methods, non-standard measures and functions to ensure that the blessed browser works. Therefore, what you see, what you use, is no more than a set of patched codes designed to work well or relatively well in your browser. This is one of the reasons why I don’t use Internet Explorer for personal use. I don’t want to know or personally suffer those things, I don’t want to enter sites and find out that the browser is already interpreting developers’ patches in order to satisfy me, I don’t want a browser that cannot understand a standard code, I don’t want to work at “half speed”, and that’s all there is.

Not even Robert Scoble, Microsoft’s blogger par excellence, uses Internet Explorer.

Nearly four months have elapsed, and all I can say is that the right decision was made. Rails is an impressive development platform. It enables developers and designers to work with the same code. It has one of the most active developer communities at present and it helps us to have an easy-to-understand compact code. Lastly, it also has a very important advantage for a start-up: it makes us terribly dynamic, distinguishing us positively from other possible competitors. The bigger they are, the better for us.

However, things are slowly changing. The other day I went to change my address at the Spanish Social Security. The good man served me with just a small smile on his face and starting hitting the keys to search for me in the database. I looked at his screen as usual and I couldn’t believe my eyes: MS-DOS 6. I told myself this would never happen to me. It’s really incredible how today’s technology is squandered. There is so much good, extensible technology that it’s hard to believe that we can’t capitalize on it. At least Spain – where I live – is on the way to digitalization, taking advantage of the digital ID card and digital signatures, compared with other European countries which see this as a contra-weapon. I believe that the idea is fantastic and, being Argentinean, I was even more surprised to read that, in my country, agreements have been signed for a «paperless» (sic) government. Undoubtedly, no-one who’s gone through the whole rigmarole in Argentina knows what it’s like to wait months for something that can take a few minutes.

Fernández took advantage of the event to announce the implementation of the Digital Signature Law that would “help to eliminate the paper and forms in transactions, thus reducing time and costs.”

The paper gets yellow and corroded and is devoured by termites after a while. Trips to city hall cost money. The time I have to wait until I’m served is not remunerated. I have to accept the civil servant’s bad mood and the truth is I don’t like that very much. I want to update my census registration with three clicks. I want to change my address in 10 minutes during the morning while I breakfast. I want to sign a contract without moving from my office and that the other party can legally sign in the next 5 minutes. I ask: why are we subjected to this? Do we have to sign by hand, wait in endless queues, live with MS-DOS and with contracts that come and go by mail?
That’s enough paper.