My personal take on DotNetPanel’s decision to go open-source
Over the past month I’ve been asked several times by vendors, colleagues and friends “What do I think about DotNetPanel deciding to go open source as WebsitePanel?” Well… I think it’s FANTASTIC! There I said it!
I’ve evaluated DotNetPanel in the past and found it to be very powerful, full of features and definitely had potential. I even own licenses that I have up and running in a lab.
5/11/10 UPDATE: Before we get started, I thought it would be interesting to have ohloh review the websitepanel project on Sourceforge and report on it. Here’s the results of that:
- 533,549 lines of code
- 142 Person Years to develop this code (estimated)
- with an average salary of: $55,000/year it would cost $7,822,306 to recreate this project.
You can click the image below to see just what ohloh sees/says about the project.
But in my opinion DNP had a few things working against it:
1. It lacked a sizeable Windows Hosting company to adopt it (someone of our size or larger). In speaking with the DotNetPanel folks this was pointed out to us a few times that most customers are smaller shops that manage only a few servers with it.
2. The company wasn’t large. Although small, agile and extremely talented I had concerns about support and it’s future as a product offering. In the past few years we’ve seen too many good hosting control panel companies take a different yet similar path to that of DotNetPanel.
3. I didn’t think the business model could really work long term. I don’t see how you’re going to cover your development costs when virtually giving away licenses for $10/month.
DNP did have some great things working for it though:
1. Most of their competitors were acquired by Parallels only to be put out to pasture to slowly run their course and eventually migrate the customer base to Plesk.
2. It came along at a time that HELM4 was launching and was very similar to HELM4. It offered a lot of the same features and functionality of HELM4 (and lacked some others). It had the potential to be a contender and I think they really benefited by Parallels acquiring all of the other control panels.
3. They were blazing new paths. They had a very nice setup for Hyper-V, support for Exchange and support for Sharepoint. At the prices this app was offered there was nothing that was going to ever come close to them. (This was also another fatal flaw though, they really should have followed Parallels, Ensim and the handful of other ISVs that build solutions around Exchange and Sharepoint at a premium.)
4. It was a distributed architecture.
5. It was XML web services based.
6. You could create a front-end for it in PHP ( I read recently that the majority of Hosting Control panels are based on PHP, think about that.. You could easily integrate DotNetPanel into your PHP based control panel).
7. Their company was fast to adopt and release support for new technology, Be it the Web Application Gallery from Microsoft, Sharepoint 2010 beta, Exchange 2010 beta or the fact that it was the first control panel available for Hyper-V!
8. The team. Feodor is awesome. I’ve had the opportunity to meet him a couple times in the past and he’s a very nice guy and you can tell he’s extremely intelligent.
9. They weren’t afraid to release bug fixes frequently and fast.
So with 2X good versus bad on the Jess list, why would they go open source?
In business you often have to make difficult decisions. Sometimes you’ll make a decision that you know will hurt some of your clients but ultimately it will be in your best interest, that of your own business and often that of the majority of your customers. So you bite the bullet and make the call.
Why do you make calls like that? They all come down to money. There’s really no stats out there saying how many installs of DotNetPanel were in play or what their revenue really was. But let’s say they had 200 monthly customers each with 10 servers online and running. Each of these customers would need a standard license at $30/month (that’s $6000/month in revenue) each of their servers would be licensed at $10/month (so that’s 20,000 in monthly revenue). Their annual revenue would have been $312,000 based on that. I was told by my rep there when I was evaluating the company that they had only 10 people. Look at the development man hours it took to write as much code as went into this application, think about the costs associated with all of those employees and quickly you start to come to realize profitable or not that their company really couldn’t have been grossing the millions and millions one would think they were and I don’t know of anyone that’s in business to just squeak by with a modest living. There’s too many headaches, liabilities and pains to do it for nothing.
So my own “conspiracy theory” is this: “the company probably wasn’t paying off in the way Feodor had hoped. He looked at his options and competition and decided the best way to really take the application forward and reduce his overhead was to go opensource!”. Plus this option would allow him to continue to work on the project and guide it and .. if he so desired he could still provide commercial offerings off of it.
But what’s the benefit of giving away your software for free?
Just ask SugarCRM. They seem to be making money doing it and think about some of the others out there. WordPress is free, they make money? And for you “Well that doesn’t work with .NET applications” Hrm.. DotNetNuke? There’s always others too.. Redhat?
You’ll gain a great deal of developers now working on your code and contributing back to it at no cost. You’ll get new features submitted and an army of QA testers out there for you and if you manage it properly you’ll benefit from it all very nicely.
AND.. There’s really only one company out there today properly positioned to support the application anyway. Yeap! CHA-CHING!!!
But I spent thousands on their licenses and now they are useless!? Cries the sissy in the back of the room.
Yes, the community has their panties all in a twist about this crying “I spent thousands on licenses for this recently!” and wondering “Well who’s going to support my application?”
First I seriously doubt you were spending thousands and if you were you were probably a handful of people doing it. Think about that…
Those of you wondering what’s going to happen to support.. you’re all forgetting .. well isn’t this an opportunity for the properly positioned company? What about a premium level of product? The way these guys were pumping out code there’s nothing to say they don’t have other features/modules they are looking to release as paid only versions.. Maybe they do a community build and then an enterprise build?
So I’d stop whining and wait to see what develops. They’re still supporting you and your customers during this transition so just wait it out..
Now let’s look back at History..
The company was small and probably wasn’t getting bankrolled by anyone and if they were they probably were burning cash. The companies that have come before them (WHA, Ensim, etc) all had superior products in one way or another to that of Parallels own product but sold out to Parallels not because it was going to get them into Forbes list of billionaires but because it was an opportunity to make a little more money than they were making and still have their customers supported. Are you really going to get rich by selling something that took tons of man hours to build, requires an insane amount of support resources and in a very competitive market at just $10/month? Not if you don’t have a way to farm those customers and grow them into larger revenue plans and products. And yeah I wrote that and said “Hrm.. Web hosting?” but the difference is we do have a growth path for our clients. This is really where DNP could have done better with the exchange, sharepoint and hyper-v offering and in my opinion should have been charging a per unit per month type of royalties license to their customers).
But who’s behind all of this? Everyone Smells Microsoft?
There’s all kinds of rumors out there floating around about DotNetPanel and the most popular is that Microsoft bought it… And although I agree this happening was a great turn of events for Microsoft, I really don’t know and I don’t care. No one has come out to say that Microsoft did acquire it and until either Feodor or Microsoft steps forward to confirm that I’d just stop spreading rumors and fishing for information. And does it really matter? I mean does it really matter? At least not to my business and if your business is so concerned about Microsoft buying your competitor then maybe you need to rethink your business model anyway!
So who’s really behind it? In the end, it’s Feodor and like I said at the beginning sometimes in business you have to make difficult decisions that ultimately you know in your gut are what’s best for you and your business.
And while we’re on it, I don’t think Feodor is the second coming of Jesus Christ so stop hating him for doing what was in his own best interest .. you wouldn’t die for his sins and you shouldn’t expect him to do so for yours.
Well What’s AppliedI going to do?
This is the next question I get asked after the first question. Well, I had already downloaded the code and had it compiled just hours after it was announced as open-source. I brought up a new lab using it and am fully evaluating it with fresh eyes. I don’t think it’s as mature as Plesk but I think it has a lot to offer and could become a great project but it wasn’t a perfect for me when it was DNP and it’s not a perfect fit now.. Though with full access the source code that changes things now doesn’t it..
Hey ALL THE BEST TO FEODOR and much success to him and the members of his team. I think going open-source is the boldest of moves a company could make and I applaud him for the decision. And for the record, I hope for him his decision to go open-source did get bankrolled by millions of dollars of Microsoft’s money and I hope he’s drinking scotch, lighting cigars with $100 bills and staring at his new Lamborghini with a never ending maniacal laugh.. But I doubt that’s really happening..