Free VHDL plugin for Eclipse?

Posted Tue, 2010-04-06 13:00 by philippe.faes (Sigasi)
Last month, we released our VHDL IDE as an Eclipse plugin. Before that, it was only available as a standalone product. There were two main reasons for releasing the VHDL Eclipse plugin version. First, many users requested features that we would never implement, such as support for C, PHP or ClearCase. Of course, we could integrate existing Eclipse plugins, but we did not want to overload our standalone product. Instead, we want the standalone VHDL IDE to be lean enough, so that we can support all the plugins we package. Releasing the VHDL Eclipse plugin version allows users to customize their IDE in any way.
The second reason for releasing as an Eclipse plugin was that the Eclipse user community is a natural target audience for our tool. Engineers that have previously developed (embedded) software on Eclipse, will likely also look for an Eclipse-based solution for VHDL development.
Now, a few weeks after launching the VHDL Eclipse plugin, we are seriously considering opening up even more towards the Eclipse community.
Today, you have to sign up on our website before you can install the plugin. Each user gets a customized update site. This is probably a bit of a security overkill. We think it would be better if anybody could download and install our plugin without first signing up on our website.
But there you have a few problems: What happens if the user does not enter a valid license key to try or buy the product? Should the plugin just be disabled until a the valid license is entered, forcing the user to visit our website and sign up for a license key? There must be a smoother path.
So here is what we were thinking: suppose that the plugin, without a valid license key, behaves like a normal, traditional VHDL editor. With syntax highlighting and code templates, but without syntax checking, without hierarchy view, without navigation and refactoring support. Whoever wants to use this editor, can do so forever, without spending any money. Users that want to experience intelligent VHDL support of the complete VHDL plugin can sign up for a free 2-week evaluation or they can purchase a license.
The free VHDL plugin would not have any unique features. It would be about as good as any of general-purpose editors out there, but it would be nicely integrated with your existing Eclipse IDE. So you would have basic VHDL support and all of the benefits of version control, issue tracking and integration with C/C++ or any other language you need.
We're still working out some technical details, but we feel this may be the next natural step for Sigasi to take. What do you think? Let us know in the comments, or on twitter.

Update We now have a Starter Edition that you can use free of charge, forever.








Comments

Of course, I am biased and

by Jakob Eriksson (not verified) - 2010-04-07 08:39

Of course, I am biased and push my own beliefs and act in my own interests. However, I feel you are missing out on an opportunity.

By taking one calculated risk, maybe you can avoid another. Let me explain:

The free product you propose will be a natural promotion of your full product. But once you create a loyal following for the free product, you have created a market for more - for free. Let's face it - at any given time the majority will not be paying. (Any individual may later do.) This majority will get used to your free functionality and expect it for free in their tool (Eclipse).

The risk you run is this: somebody else will make an open source tool. At first it will not be as good as yours, but it will be part of Eclipse from the beginning. Slowly, it will evolve features, until it is not only as good as your free product, but better. The open source product will displace your free offer, so you will loose the promotion you had. The open source tool will also be better and better from each release, until it rivals your professional offering. You will still have your loyal customers, but can not expand very much further.

How will you counter this scenario?

No problem - you will add more features to your free of charge product. You might just as well. After all, you are constantly adding features to your professional product, right? So you may just as well let some of the features in, slowly, into your free of charge product without losing too much.

This is all hyperbole, of course. It could be true however. The proposal I am making is this:

You might just as well release the free version not only free of charge, but also open source. Then as you update the professional version, you also let some of the features into the open source version. This requires discipline and good judgment. What's the benefit for you?

If you maintain integrity and a good reputation in the community, you will never be overtaken by an open source alternative, because you would be the open source alternative!

You also mentioned distribution. If the free version is open source, you can get it accepted in Eclipse, and by extension into Ubuntu and Fedora. This is powerful promotion.

There could also be a focus internally at Sigasi on helping paying customers adapt the professional tool to their specific corporate environment - maybe integrate it with their in house EDA tools or whatever. This builds loyalty like nothing else. The generic features useful to other customers will migrate back into the professional "shrinkwrap" product and eventually into the open source product.

A competitor would have a hard time competing with the ubiquitous presence of the free offer and the steady increase of value and features from paying customers.

But it is your money and your decisions to make. You asked for input or I would not have been so bold.

Hi Jakob, Thanks for your

by philippe.faes (Sigasi) - 2010-04-13 08:58

Hi Jakob,

Thanks for your opinion. We have considered open source as an option. Here are some thoughts that made us decide not to move to open source at this time.

  • It is a considerable effort to split the product into an open-source part and a proprietary part.
  • Sadly, there are very few EDA open source projects. If you'd look at project health metrics (number of commits, number of questions on their mailing lists, number issues fixed, etc.) you'd see that open source EDA projects have small communities and move very slowly. This makes me believe that not many people have the skills, the legal ability and the will to participate in an open source EDA project.
  • A company that releases its own software as open source, would expect to build a community of contributors around it. Unless there is a good open source community around the product, the company would loose more than it gains from releasing the source code.

I do not see many open source EDA projects emerging in the first few years. However, I challenge anybody to prove me wrong. I do believe that the EDA sector would benefit from some high quality open source projects that challenge the status quo.

Yes, it is a considerable

by Jakob Eriksson (not verified) - 2010-04-20 08:37
    • Yes, it is a considerable effort to split a product. I have been in the same position as you and made the same decision. Sometimes the effort can not be justified.