Bob Keeney recently posted a blog entry Is REALbasic Relevant?. It has generated some noise in the REALbasic community. Bob is known for his brutally honest assessments of things and this post is no exception. As you may know, Bob and I are both on the board of ARBP (he’s the President, I’m the Secretary). Since I’m often known for evangelizing REALbasic, I thought I would add to the healthy discussion by offering a counterpoint of sorts.
So, is REALbasic relevant? Well, the question is “to whom”? It’s certainly relevant to me and anyone else that uses it. It’s probably not entirely relevant to the development community as a whole, but then it really never has been. Unless your name is Java or the latest technology du jour from Microsoft you’re largely ignored. REALbasic is no exception there. Until recently, even Objective-C/Cocoa was irrelevant.
Cocoa
As Bob notes, he’s talked to lots of developers that have recently started looking at Objective-C/Cocoa, but I don’t think that makes REALbasic less relevant. In most cases, it’s just smart business for the developer. If you’re looking to create a consumer Mac-only application, Cocoa is your shortest path to providing the lickable UI that your Mac users will expect. And add to the mix the flurry of people jumping on iPhone development, which primarily can only be done using Cocoa1 and you’ve got even more people looking into it.
REALbasic is absolutely wonderful for creating niche applications. If you are looking to sell a niche application, then you’ll likely want to have it work on all platforms in order to get as much out of your small market as possible. REALbasic makes that easy. Since your application will likely fill a specific business need that is not being met elsewhere, the UI of it won’t be the deciding factor. You can still make your UI look nice with REALbasic, but it won’t have all the latest gadgets for each OS platform. However, your users just won’t care.
And if you create custom database applications for businesses, I doubt they’d want you to take the extra time to create a Cocoa-licous OS X application. As slick as Cocoa and Objective-C might be, there is no denying that REALbasic (and other tools) would allow you to create their app faster.
Database Servers
Speaking of databases, there have been plenty in the community who feel that REAL Server (formally REAL SQL Server) is a waste of REAL’s time and resources. The world doesn’t need another database server, they say. I have to disagree. There are already a lot of database servers out there:
- Oracle
- MS SQL Server
- MySQL
- PostgreSQL
- OpenBase SQL
- Valentina Office Server
- DB2
- Firebird
- Informix
I’ve used most of these and I’m sure there are many more I’m missing. My point is that the database server market is incredibly fragmented. A fragmented market is often the best type of market to enter. And REAL Server has a pretty significant differentiation compared to all these other database servers: it is simple to use. Simplicity is significant. The process of installing any of the above database servers is often not easy. And even if that’s just a one-time task for your or your client, the job of administering it can often be far more onerous. When I choose a database server for a client, it is not based on what I prefer to use, it is based on what will work best for them.
Sometimes (or maybe even frequently, depending on your client base) REAL Server will be a terrible choice; after all it is missing some pretty significant features2. But for a small business that does not have a database server or little to no technical expertise, it may be a really good choice.
Personally, I’m not fond of REAL Server as the new name (or the old name of REAL SQL Server for that matter). To me, it doesn’t denote what it is. Is it a database server or an app server? More importantly, how are developers to know that it is built on SQLite and not an all-new creation that has no history or credibility? SQLite is a powerful brand, so leveraging it would be helpful for REAL, I would think. I’d prefer the name REAL SQLite Server. That alone would help the many people using SQLite today (and not just with REALbasic) realize that there is a server to easily upgrade to. And it would likely result in better Google search results.
A brief aside related to databases: I will say that one thing that everyone will agree on is that REALbasic needs the ability to visually create reports. Database applications almost always heavily rely on reports. Right now, it is more difficult than it needs to be to create these reports in REALbasic. On-Target Reports is very restricted and has some annoying bugs. The new Report Designer included with the Valentina 4 beta looks promising, but it only works with Valentina databases.
Cross-Platform Development Tools
Lastly, as far other cross-platform tools killing REALbasic, I think that is unlikely for the foreseeable future. Cocotron looks interesting, but I’m not a big fan of Objective-C. Far too many open-source projects like this have died on the vine and it’s not like this project is going to get any support from Apple or Microsoft. Apple has zero incentive to see something like this succeed. I’m sure Microsoft views it as irrelevant.
I’ve been following Mono for years. It has special appeal to me since I’ve worked extensively with Microsoft .NET. But even .NET diehards will tell you that .NET is not really being used much for desktop applications; it’s big draw is web apps. Add to that that there is still no clear cross-platform UI library (Winforms/WPF works on Windows, but not so much on Linux and OS X; GTK is good on Linux, so-so on OS X and poor on Windows) that it’s not really a viable cross-platform solution just yet. Heck, the MonoDevelop IDE only recently started working on OS X at all. And if you’re really thinking of using something like Mono, you should just be looking at Java instead. It has been out forever and already works for creating cross-platform apps. And it’s insanely complex, just like Mono.
As for Lazarus, if you think REALbasic is not relevant, then Pascal is even less relevant. I like Pascal, though, and keep meaning to try this out.
One that Bob didn’t mention is Adobe Air, which I think looks interesting. However, from what I’ve seen so far, apps built with it don’t look platform native at all.
Wrapping Up
I know it’s clichéd, but I advocate that developers pick the right tool for the job at hand. If you only know one tool, you’re not setting yourself up for success. Is REALbasic relevant? If you get something from it (competitive advantage, enjoyment, learning or whatever) then who cares what others think?
I have to continue to believe that REAL Software is working diligently towards improving their products. And from what I’ve seen, that is true. They could certainly use more transparency, though. I’d love to see an official company blog, for example.
1 I am aware that some folks have been able to create iPhone apps using Mono and sometimes iPhone web apps can work well
2 To name just a few: referential integrity, stored procedures, partitioning, replication, terabyte databases
8 Responses to “The Relevance of REALbasic”
I would absolutely support more transparency. I think more insight on what’s going on at RS would be great for the community. That’s what I really enjoyed about Aaron’s blog. Even if it wasn’t an official RS blog, you got some insight on what people were working on and what were the challenges and forces behind some design decisions.
Yes, I’m already missing the insight that was provided by Aaron’s blog.
Thanks for the kudos.
I wish I could continue to provide insight into REALbasic on my blog, but honestly, I’ve sort of lost the drive to do so. However, I suspect that as I learn more about 4D, I’ll get a new drive. 
In terms of whether RB is relevant or not, my answer hasn’t changed in seven years: it depends entirely on the person’s needs. If RB fits the bill for a particular project, then it’s relevant. If it doesn’t fit the need, then it’s not.
@Aaron: Looking forward to your ramblings about 4D then
I’ve until recently used a software which I think was written using 4D but as it was absolutely not straightforward to use (not to say almost unusuable) I’ve replaced it with an application written in RB (and not a self-made one). However, the usability of that application probably has nothing to do with 4D itself.
another great basic is purebasic !
work on linux,mac,windows
http://www.purebasic.com
RS has just announced their blog:
http://forums.realsoftware.com/viewtopic.php?f=11&t=26256
It can be found on:
http://realsoftwareinc.blogspot.com/
Very nice!
That’s great new, Jef! Thanks for letting me know.
Hi Paul,
I hope that REALbasic gains more relevance. I´d try to convince my customers to let me solve their problems with REALbasic since about 5 years (or longer). One customer finally did and now is happy to have done so. I am happy, too, because we are carrying a project that would have been much more expensive with any other IDE and Language I know. But in Germany, with REALbasic you are kind of a maverick or don´t have the balls to produce substantial code like with java or C++.
With REALbasic you are very effective, but a little isolated and at the schoolyard the others point with their fingers on you. Let´s call that relevance.
Jens