Why Rich Internet Applications are better than AJAX

I recently spent some time in a grindhouse development department learning Flex as well as supporting some basic AJAX features. I’ve also been working with Silverlight for over a year as a part of my Textfyre work. In the past, I’ve worked on projects that used some DHTML. I’m familiar with RESTful web services, jQuery, HttpWebRequest, and all of the other aspects of building highly interactive web apps. I’ve never done any real Java work, but I have played around with the technology enough to know that you can do decent things with Java (although the visuals have always left something to be desired).

Throughout these efforts I’ve always kept an eye what businesses might consider the most cost-effective solution for scaling their applications, whether the application is for internal use, external use, or a hybrid. Google took the lead in developing applications with AJAX when they introduced the nicer features of gmail.com and Google Maps. This was the birth of Web 2.0 and I would say we’re still in the middle of that era. A lot of companies have adapted to AJAX using jQuery and other toolkits. In many cases, using jQuery to liven up a website is very successful.

Then there are websites that are much more than just a website. We call these web applications and they tend to try to do many or all of the things that a desktop or “fat client” would do. These applications will load large amounts of related data, like customers, orders, transactions, and more.

A few years ago, our PC’s weren’t capable of handling this kind of scenario. It wasn’t the browser, but the lack of memory, disk space, and bandwidth. Today, most people have access to high speed Internet access. Most people have newer computers with more memory and more disk space. Certainly many businesses have either adopted these kinds of computers or will within the next year or two (in their move from XP to Windows 7).

So today we have the platform and the ability to make highly interactive and data intensive web applications. But the first question anyone should ask is, what are the requirements? I’ve seen a lot of applications ported from perfectly sound desktop applications (with deployment headaches, but still, they worked) to web applications. Deployment issues aside, there was no other requirement for the application to be built in web technologies. In many cases, the users lost many features and flexibility when their apps were webified. The process of moving a complex desktop application to the web paradigm is very expensive. We now have people called “information architects” that take very sound desktop user experiences and port them to web technologies. Why do we need these people. Because building smart web applications is vastly more complicated than building smart desktop applications.

There certainly are benefits to having an application webified. In a desktop application you might have a very strict set of menus and options or a single path to each feature. In a web application, you can develop multiple paths to features with varying usage scenarios. Some of this is possible in a desktop application too, but a web app offers these types of options through links and is much more seamless.

Benefits aside, web applications are very complex applications to build and moreso, to maintain. To allieviate the maintenance costs of building web applications, tools have been developed that make creating and debugging most applications very easy. Inline debuggers allow us to literally step through every line of code, see the stack trace, review element values, and resolve issues without throwing darts blindfolded. Debugging used to be a much higher level skillset. It’s still an art and skill, but with the tools today, it’s much more widely available to the average programmer. With an exception. AJAX.

AJAX is great, as I said before, for making websites behave nicely. But when you take the technology of AJAX and marry it to a complex web application, you’re exponentially adding complexity and cost to your application development and maintenance budget. The tools for debugging AJAX are not simple. In many cases a developer will resort to using Fiddler (a tool that watches HTML transactions) and arcane alert(‘here I am!’); statements haphazardly placed within the application’s javascript. This is a nightmare scenario for any developer. The more AJAX you implement in a web application, the more complex the interactions get and debugging capabilities rapidly deteriorate. And this is if your codebase is developed well and is highly readable. If the code was slapped together in a rush, forget it. You may never completely understand what the code is trying to do and it could be safer to do an organ transplant instead of minimally invasive surgery.

With any problem, someone inevitably sees it and develops a solution. Oddly enough, Sun developed a solution before the problem existed with Java technology. The problem is that Java made more of a name for itself in the server world and standard web application development. The client side features of Java were left to cute games and scientific demonstrations. You can blame a little of this on Sun for so desperately trying to make Java pure across platforms when they should have recognized the value of implementing a VM->Native compiler for each platform (Sun, Mac, Windows, Unix, etc). Instead, Microsoft saw the potential and created .NET, not necessarily to be cross-platform, but to enable managed code, garbage collection, for client-side and server-side development.

Eventually Sun realized the potential for Rich Internet Applications and has been actively marketing related solutions. But so has Adobe with Flex and Microsoft with Silverlight. These three platforms enable cross-platform application development that does not rely on web technolgies like html, css, and AJAX. These technologies offer front to back development environments with very robust and clear layout engines (arguing that CSS is hardly clear although it is robust), strong debugging tools, and deep integration with web services.

This is where application development is headed and it’s primarily because CIO’s have started to quantify the cost of developing web applications. We made enormous strides in resolving deployment problems (10 years ago we had to package software to be installed automatically on thousands of PC’s upon login and if there was a problem or virus, well, it was very painful), but at the cost of lost features and very expensive maintenance.

We now have the ability to deploy applications that more secure ane vastly more stable than before. Managed code, which is used in all three platforms, is the key to the stability issue. Security has become a primary focus within Microsoft and no one would argue that their systems and applications have improved dramatically since the early XP days of major viruses appearing daily.

We can now develop middleware that is adaptable and maintainable at a low cost. We now have the ability to develop disconnected presentation layers that can manage large datasets, but is also easily developed and maintained. It looks pretty too, if you have good designers.

If you look at new development requirements coming out of the major job sites like Monster.com, CareerBuilder.com, Dice.com, you’ll see that Silverlight and Flex have taken off. I’m not sure why Java hasn’t, but I assume some companies will adopt the Java desktop model too.

Throw away your CSS and AJAX libraries. You’ll save time, money, and your users and your infrastructure staff will love you. So will your shareholders.

The Perils of Consulting Part 2 – The Greatest Job Ever

Updated on 12-3-2009.

It all started out when a staffing firm called me up in late 2008 about a two year old social networking website, OneDegree.Com. I took a look at the site and thought they had a pretty good chance of competing with Facebook and MySpace. The role was for Sr. Architect and had a lot of potential in my career path. I told them I was interested and to set up a face-to-face.

I met with the two directors, Ted Stenstrom and Jeff Baker, at a coffee shop in Chicago. Ted and I had a pretty good rapport with where the site could go and what revenue streams would be tracked. They said they were “fully funded” and had a million dollars in the bank, but things were still pretty early in the development stage. They were missing some of the basic pieces of a company, like an office, basic staff for HR, etc, but all that would get put together in six months.

After a couple of weeks, we had a second meeting, at another coffee shop, and nailed down compensation and parameters for a contract. The numbers, in a severe economic downturn, were very good. I wasn’t going to get an offer like this anywhere else (for good reason, but I’ll get to that later). With compensation nailed down, we talked about where the development team would be located. I suggested opening an office near my house in the western suburbs of Chicago and they thought that was a great idea. I suggested a Regus office and they said they already had two Regus offices and this “fit into their model”.

So I made arrangements to move into an office in Lisle, Illinois at the Regus office. I was able to get a great corner space with two closed-door offices and a “bullpen” where we could put two other developers, the server, and the fax/printer. Having space in Regus is like being in a country club for small businesses. It’s really very comfortable.

While procuring the office, I was also tasked with identifying what kinds of talent would be needed to re-develop their website. Oh, about the website, a little background…

The original website, the one that is still online (with minor adjustments), is a static site. It’s really a design prototype with almost no flexibility or business intelligence built in. Some of the features don’t even work very well under load. We couldn’t even put the site behind a load balancer because it just wasn’t built to scale. It was developed by Manifest Digital and a smaller company called Tecture. I believe there were some issues in what exactly was being developed, but I don’t know enough to say. I do know that Tecture had a legal battle with OneDegree over payment and eventually settled. I do not know the details of the disagreement.

In any case, I ignored all of the warning signs. I was being offered great pay, my own office, my own development team, hardware and software of my own choosing. It was just too good to pass up…..and too good to be true.

In order to find people to help out, I worked with a staffing firm I know and respect, SPR in Chicago. I also asked a colleage, Jack, if he wanted to come on board directly. Jack waffled for a bit, but then said yes. The rate OneDegree agreed to pay him was too good to pass up. Also too good to be true. Through SPR I met Heath and Mark. I needed only one senior developer, but SPR convinced me to take both. I then heard that one of the guys from Insurance Auto Auctions was looking for work, Mayank. I hooked Mayank up with SPR and they hired him and placed him on my team.

I was in the office by myself for the first week and procured the hardware. We ordered about $23,000 worth of equipment from Dell which included a server that could host multiple virtual servers and laptops for the developers and myself. During this process, there was an issue with payment. Ted didn’t have a credit card to pay for the hardware and no one else seemed to have one either. I would have expected someone to have a corporate card for the business, but apparently not. We actually were on the phone and trading e-mails about sending a paper check. Then there was an issue of when the check arrived, and then waiting ten days for the check to clear.

Meanwhile, Mark, Heath, Mayank, and Jack started and sat on their hands for several weeks. We got the hardware finally and put the development environment together. I hired another colleague, Kevin, on a short-term basis to put the network together for us.

For software, I partnered with Microsoft through their BizSpark program. All start-ups, less than three years old, with less than a million dollars in revenue, can procure MSDN licenses for their development processes for free. We put together Windows Server 2008 virtual servers, implemented an Active Directory server, a SharePoint server, and a Team Foundation Server. It was a pretty sweet environment.

At this point I had not signed my contract yet and we were going back and forth on the language. OneDegree had not even put an initial contract together for Jack and all e-mails about it were deflected with statements saying the lawyer was ”kind of busy with other things”.

I was expecting my first check on February 15th and ended up picking it up from the accounting firm Seldon Fox later the following week. I was expecting my second check at the end of the month. It didn’t show up anywhere and Ted told me to check my mail box at home and at my business box (at a UPS store). Jack wasn’t seeing a check either. It then came to light that no one knew where the check was and Seldon Fox said I needed to speak to Tommy Davis, the owner of OneDegree, to find out the details.

So I called Tommy and he apologized for the delay and said they were out of money. But they were waiting on new funding to come in “any day”. He assured me the next round of funding, anywhere from $500,000 to millions of dollars would be in at any moment and it was just taking more time than he expected.

I said I needed to get paid. So he had a colleague or friend, Joe Blake, loan OneDegree enough money to pay Jack and I for two weeks of our time. The other developers were getting paid through SPR, but SPR wasn’t getting paid either. I found out soon after this that Regus wasn’t getting paid and we could eventually get evicted.

But in talking to Tommy and hearing his pleas and wanting to keep this job and knowing that there wasn’t any other work (consulting work had dried up in the entire Chicago area and probably the whole country) I made a call. I called a colleage of mine who I know has started businesses and knows his way around this world. I put him in touch with Tommy Davis and I agreed to borrow $20,000 from my friend and Tommy agreed to pay a sweetener of $2,000. This was written into the contract that remained unsigned by OneDegree.

I picked up the check from my friend and deposited it. I also had lunch on St. Patrick’s Day at Houlihan’s with Tommy Davis. We chatted and he seemed like a decent guy. I was probably more concerned with getting paid than truly evaluating Tommy’s character though. Tommy came back to the office and we signed my contract. He did not offer to sign a contract with Jack.

Part of the new deal was for Tommy to report to me every few days on the progress of the investment pipeline. He did this twice and then stopped. The document had a lot of nice words about “any day” and “this one is ready to pop” and similar things. The two copies of the document were essentially the same and probably why he stopped sending them to me. There was no real progress in acquiring funding. The investors, for whatever reason, were backing off their commitments.

So in mid April, still not being paid, Jack without a contract, I loaned Jack $4,800 as a good faith measure to keep the team together.

I was also invited to a board meeting with Ted, Jeff, Jason Benedict from Manifest, Kevin Knaul, Joe Blake, and Tim Finnegan. Note that Tommy Davis, the owner, was _not_ at this meeting. The goal of the meeting was to discuss a short-term change in the business model to make OneDegree a jobs aggregation website, make a few bucks to “prove the model” and then this would lead to the “big investors”. There was no talk of funding, no one wanted to talk about when I was going to get payments for my invoices, and no one wanted to talk about Jack Marchetti not having a contract and not getting paid.

But I brought it up and pushed pretty hard. This got everyone’s hackles up, especially when I mentioned that I might bring lawyers into the process. That really pissed them off. At this point, I had probably lost all confidence from these guys. I went to a meeting for virtual job fairs with Ted and Jeff a few days later and the mood was cold if not, “why are you even here?”

I was determined not to let them scare me off. I wanted to see this thing through, but my days were numbered from this point forward.

Ted then started telling me that they may have to break up the development team and use offshore talent through Manifest or Indecon, Kevin’s company. So I started that process by letting Mark go. This pissed off Heath, Mayank, and Jack, but they didn’t know that Ted was pushing for this and expected to let one more person go.

Through April and the first half of May we kept working on the new website. I had developed the architecture along with Mark’s help and we were making a lot of great progress on building a scalable replacement for the static site you see today. I had a new data model and was building data transfer scripts to move the existing data over to the new model. Then Ted calls me. Up until this point he had been very confident that the money would be in “any day”. But this time it was different. He said it could be months before they see funding. He asked what I was going to do. I said I’d hang on part-time and look for contract work. When funding came in, I’d be back on full-time immediately.

He didn’t like that idea and it was clear he wanted to end the relationship. I pointed out in my contract that they had to give me 30 days notice and that I had no intention of quitting. The following day he sent an e-mail with a formal 30 day notice. I worked from home for the next four weeks, having moved out of the Regus office. A few weeks after I moved out, they moved the developers and hardware to SPR’s offices.

I helped Jack find a paying job through another staffing firm and SPR pulled the cord on free development. They placed Mayank at another client and Heath found a project somewhere else.

At the end of all of this, they owe me $82,000 and they owe Jack $26,000 and he still doesn’t have any paper to show he worked at OneDegree, other than e-mails. they never signed a contract or any employment agreement. Both of our debts are in collections. I have on record Keven Knaul saying they plan to pay Jack but they probably won’t pay me.

I have no idea how much they owe SPR, but I assume it’s a lot. Probably well over $100k.

We’ve had some run in’s over their domain since they had placed me in charge of its registration. I also had control of their gmail accounts, but all that’s been returned. The hardware sits dormant in SPR’s offices and the company is in a holding pattern.

In the end, this entire fiasco has probably cost me $200,000. If I never see the money they owe me and I have to make up that money somehow, and I was out of work (still am) for months without getting paid….it comes out to about $200,000 of lost revenue.

In hindsight, there were warning signs all over the place. I wanted the job so badly, for personal and professional reasons, that I was blind to these warning signals.

I won’t ever be blinded again.

Updated on 9-16-2009

From the collections agency…

* * * *

Dear David,

The following is a copy of our diary notes from my conversation with Kevin Knaul today.  I was directed to him by Ted. The notes are self explanatory.  Currently the company condition does not look good:

Kevin called said he does not work for the company said he was a consultant, said basically the company does not have any employees all worked as consultants.  He said Tommy Davis is the owner and majority shareholder, also said the business is broke.  He said he is due to have a meeting with Tommy later this week.  He said Tommy wants him to “re-engage” with the company and try to raise some capital.  He said Tommy, himself, is not a very “in-vestible guy”  Kevin has not made the decision whether to begin working for him again because he knows at this time it would be for free…but if he does securing investors may take some time, he said currently they have a “static” site with no marketing…said he thinks he has a good idea and may consider working for him if he can set it up to get some sort of equity position if the company becomes profitable again.  He said he has been out of touch since December so he has to become familiar with exactly where the company currently stands but believes it was always their intention to pay John Marchetti.  Said David Cornelson may be a tougher story as their is some bad blood due to the fact that David took their website down or redirected during a period when investors were looking at him, said they lost confidence since Tommy could not even keep his website up. Told him you certainly can’t blame David when he is holding a receivable of this size!  Kevin said if he gets involved again he may be able to smooth things over.  Told Kevin will touch base in one week to see what direction they are going after the meeting, said it would be practically impossible for us to touch base with Tommy directly as he believes he is “hunkered” down in his garage somewhere trying to avoid even his personal creditors. 

* * * *

I think that note speaks for itself. Ted is now denying any knowledge of Jack Marchetti working for OneDegree, asking for executed documents to prove otherwise.

About a five weeks ago, I was contacted by Kevin Knaul at about 9pm at night. He mentioned that he had found this blog entry and was concerned that investors of OneDegree might look at it as hindering their business practices. I find this ironic because OneDegree has severely hindered my business practices by not paying me. In any case, we talked for a bit about the past dealings with Tommy Davis and Ted Stenstrom. Kevin said Tommy, although still a major shareholder, was not managing the company. Kevin said he and Tim Finnegan had taken over the direction of the company and were working on bridge financing and funding. I took this as a good sign and said I would happily make “amends” and help them in any way that I could (outside of actually doing any work without getting paid!). I also agreed to take this blog down.

Well I had two update phone calls with Kevin and those were cordial, but there was no progress. The last call was three weeks ago. I sent a few e-mails, left a voice mail, but Kevin hasn’t replied or returned my call.

In lieu of the non-communications, the blog is back up. I really hope that OneDegree gets everything together, but this doesn’t change the fact that they owe me $82,000, they owe Jack $26,000 (of which he owes me $4,800), and they owe Kevin $3,400. If they settled these debts, we could move on.

* * * *

Updated 11/24/2009, almost exactly 10 months from my start date at OneDegree.Com.

I just noticed the website is down and Kevin Knaul has removed all reference to OneDegree from his LinkedIn.Com profile. So that would seem to be the end of it. This blog stays up as a warning to anyone else doing business with an Internet start-up in this economy. Don’t trust anyone. Do your due diligence. If you do sign on with one, have an escrow account setup. Have someone show you an American Express card with the company name on it (this will some amount of credit worthiness). I would ask for a financial profile from the officers of the company, even if your at the bottom of the food chain.

* * * *

Update 12/3/2009

A placeholder is back up. Wonder what the heck they’re (whoever they are) is doing. I suspect Tommy Davis is trying to get it back up. He probably has nothing to lose by keeping the site going until he can figure out a way to pull it all back together. That’s what I would do.

The Perils of Consulting – Part 1

In November of 2008, I was winding up a project at Insurance Auto Auctions. The project had gone extremely well and I was walking out with a letter from a director on how satisfied they were with my work. Over the course of ten months, I made increment changes to their development practices. They already had a very mature and strong development environment that included all the necessary hardware and software. No architect or developer could ever say that they weren’t given the tools at Insurance Auto Auctions.

After convincing management that we needed a dedicated build server, the first thing I implemented was a set of middle-ware templates that acted as a framework for accessing their databases. Not as robust as nHibernate or any of the common frameworks, but tooled for their environment without all of the extra features that either confuse developers or entangle projects with needless work. For instance, the templates I use do not have relations built-in. You have to wire them up on your own. For most traditional CRUD development projects, you rarely need more than Fowler’s Table Module enterprise design pattern anyway.

The second step was to implement continuous integration using Cruise Control .NET. A fairly common tool in the development world, created by Martin Fowler’s Thoughtworks consulting firm. This gave us immediate feedback on checked in code.

Thirdly, we went round and round about how to build the new system. Did we want to use traditional WebForms? We’re we okay with using the beta of ASP.NET MVC? Did we want to role our own MVP (Model View Presenter) framework or possibly do something else? In the end, we decided the simplest implementation, for various reasons, was to create a home grown MVP framework. Of course MVP is considered obsolete now and we saw what people were recommending, but we moved forward anyway. We felt the ability to cleanly write unit tests was a critical aspect. The downside of course was that we had to rewrite a number of HTML controls that come out of the box in ASP.NET. The out of box controls use ViewState and with ViewState off, we had to go it alone, so to speak. Luckily we had Cliff on the team. Any team that has some has brilliant as Cliff should feel blessed. I’m a pretty smart guy, but more suited to larger problem solving. Cliff has a knack for needling through really fiddly problems with a very elegant solution.

We of course got buy-in to implement unit tests in Visual Studio .NET 2008 Team Test. We installed this on the build server and after every check-in, all tests were executed and the build failed if any tests failed.

The business side worked on user stories, the developers had meetings to gain an understanding of each story, and the rest they say, is history. the development went very smoothly. By implementing all of the code in a single solution and by also implementing a nifty web.config connection string locator, we were able to execute deployments with the built-in Publish feature in Visual Studio. After dealing with complex and error prone deployment processes in the past, the project manager was deleriously happy with the new process.

We deployed the web application to several servers in a farm and after some fairly minor performance tuning, the system was deemed stable and in production.

In the last few weeks of the project, I was also working on securing my next position. This proved to be a very important decision and one that seemed too good to be true. Alas, that old saying is only the tip of the iceberg and will be discussed in Part 2.

Introduction

This blog will be about technology and business. I have twenty-four years experience in the computer industry, more than half of that working in the Chicago area with a lot of big companies. Firms like Accenture, Sara Lee, Chase, BP-Amoco, Cargill Financial, and GATX Rail. I’ve also worked at a few medium-sized business, and a handful of smaller shops.

I’ve also worked directly for some of these companies while others I worked at as a consultant. I’ve worked as a consultant for firms like Avanade, but most of my consulting has been as an independent, mostly through staffing firms.

In the first half of my career, I worked on DEC equipment like PDP-11’s, VAX’s and similar systems. In the mid-90’s I started working on PC’s and became entrenched in the Microsoft technology world.

The first few entries are about my last customer, who may or may not remain nameless. I’ll let my attorney decide on whether I can actually name them or if I should. The goal is to write stories (good or bad), but retain an ethical standard.

Fee free to comment and share your thoughts.

David