Hello Martin;

will u release our dream soon?

Is there any further updated release date? like in Q1 of july? or Q2?

Thanks...

First alpha during july, same as before. I'm in france at the moment but will be able to focus more on hMailServer when I get home again.

Cool have a nice trip:)

So I'll kick off my portal with 4.3; upgrading will be so funny

Thanks

It's still july

Alpha yes.

I'm looking forward impatienly

Feel a bit embarrassed but there will be some delay. I'm leaning towards releasing hMailServer 5 as closed source (still free) which means that I have to resolve some licensing issues first. :-\

No worries man take your time hMailServer 4.3 works fine and still enough.

I've never felt like I need the source code so doesn't metter for me but I forecast that if v5 will be closed source, v6 won't be free. (Which I think shouldn't) I hope I can make enough money to buy till that time

Not something I am worried about. Never really had a reason to look at the source anyway.

I'd like to take a look though eventhough I don't understand anythink Hehehee..

Hey Martin,

I have no concern with you shifting to closed source either. What I am wondering though is why? Just wondering what factors would enough to make closed source a better option for you, and a lesser extent your users?

A few different reasons..

In total, I think there may have been about 40 people who have submitted code to the project. Out of these, perhaps 4-5 or something has actually submitted something useful (the addons for SquirrelMail and the previous version of the PHPWebAdmin). The remaining submissions have only pretty much been a waste of time for me, since they have been very specific to some single users need, or just badly coded. Most users who have wanted to contribute have downloaded the source and then not done anything (after I've spent hours on helping them getting started). I understand that that is because I've never actively looked for people to help me though. But the fact is that so far, in terms of development speed I've lost more than I've gained.

Second, some companies have downloaded hMailServer and built their own commercial products on top of it in violation with the GPL license (claiming to own copyright, not distributing the source and more). So I have had to spend time on explaining to them that they are breaking the law and etc. Not that it's been that much time spent on it, but it has made me a bit de-motivated at times. It feels quite stupid to sit and write code for free when other companies break the law that way. (Each one of them claimed that thought that open source meant that the software was free for anything).

Third, a lot of companies feel that just because the software is free, they have access to unlimited support over email / telephone.. People have called me in the middle of the night requested support. People might continue to do that, but at least I'm not giving away the source code to them.. (okay, this argument doesn't make much sense.)

I will hopefully start to offer commercial support for hMailServer soon which I suspect will be easier if the source code is closed. I feel that some companies who have requested quotes on support services have believed that because it's open source, the prices for support services will be extremely low. One of the largest organization here in Europe contacted me and requested support for about 100 servers spread across more than 20 countries. They wanted to pay not more than like 250 EUR per month in total, claiming that they were a non-profit organization (even though this organization manages a budget of over 100 billion euro and supporting such an organization would most likely be a full time job). What I mean is that I find it somewhat annoying with companies and organizations who have serious amount of money but want to "ride the open source wave" and get away cheap without really contributing anything them self..

The single big downside I see with distributing hMailServer as closed source is that I will no longer be able to distribute MySQL due to license incompatibilities. I talked with MySQL AB and they wanted me to pay some EUR 10 000 a year for distributing the client library for MySQL in a closed source app. Since my budget is estimated to EUR 0, this will make the installation somewhat harder for new users..

Hopefully I don't sound to bitter now, I'm in a quite good mood

Go on Martin, I don't even understand why hMail is free? Which actually teach me a lot

Hey Martin, I feel your pain man. I've project managed open source apps before (although nothing on the scale of hMS) which I have funded with my own money (as I lack the coding skills myself), and it is really hard to keep the head up sometimes when *some* people insist on treating you like a doormat with little to no consideration that everything you are doing is voluntary. You've lasted longer than me though. I lasted about 8 months before throwing in the towel, I just couldn't put up with some of the attitudes.

However, I'm not entirely sure how going closed source is going to lessen the problems you've outlined in any great amount (with exception to those building on top of your source). From what I gather from what you wrote the main issues you see are;

1) Developer recruitment, training and retention
2) License infringement
3) Realistic support expectations

Is this correct?

If so, then I'd expect a lot more could be achieved in clearly defining a few support expectations and involving your existing community more than closing up the source. Is there any reason why the initial liaising with developers and those seeking commercial support could not be handled by members of the community other than you? Same goes for the chasing of license infringements. Once the infringement has been identified there should be no reason other members of the community should not be able to help. I know I'd be willing to help out where I can.

Also, don't under estimate the appeal of open source to companies. Especially small to medium size. I have used the "hit by a bus" scenario in several business justifications. I.e. if a closed source application provider (especially the smaller ones) should go out of business, then there is a very real business risk involved as your company can be left high and dry with no where to turn for support and the prospect of a very expensive data migration to an alternative. With open source if the maintainers should get "hit by a bus" then it is possible that someone else will fill the gap and in a worst case scenario the source code is still available. Then there is also the far more deliberate act that some closed source companies use to restrict their users choice through use of proprietary standards, which is not the case with open source.

Then again, you might have other plans for hMS entirely which is fine too ...

I just listed some examples of problems with the current way I do it. There's plenty of others as well and I think you misunderstood me a little bit. Sorry if I'm being unclear. A bit stressed at the moment and trying to explaing feelings and etc is hard over a internet forum.

About support.. There's two reasons I want to start to offer commercial support. The first one is that it would attract other users with higher demands on support (which is often fun to work with), and the second (in direct) is that I want to be able to spend more time on this project. I can only spend more time on the project if I get some financial compensation for it since I don't have the economy to work full time on this project for free. But I want the software to be free (in terms of money), hence support and other commercial services is a good thing.

I don't think I can make a living on that with the current size of the user base though. To be able to increase the number of users who use the software, I'm considering taking a leave from my job for at least half a year adding new features, improving the documentation and so on. The purpose of increasing the number of users is that hopefully more will be interested in commercial services which could then sponsor the time I work with hMailServer. Feel free to call me greedy, but if I'm going to spend like 10 months working full time on this project for free (living on savings and not getting any salary), it feels like a big personal risk if the software is open source. When I'm "done", anyone could pick the source code, make a fork of it and offer the exact same services. (It's a personal risk no matter what of course but I feel it's smaller if the source code is closed. Feels like a fun task though...)

Also, I've gotten offers from companies who have wanted to build new products on top of hMailServer and who have offered me money for it. I wouldn't mind if some company built a commercial anti spam solution on top of hMailServer, but I can't sell hMailServer source today due to the current licensing model. I'm greedy, but I think a large majority would benefit if I spent more time on this project.

I'm aware that there's a risk that a certain group of users will drop/not use hMailServer if it's only free and not open source. I'm also aware that changing to closed source might be a terrible idea and a waste of time. I also agree that using closed source is often higher risk than using open source software. To reduce those risks I'm leaning towards offer some kind of escrow-type agreement. Also, companes/individuals who I trust will not redistribute/sell the source code will probably be able to obtain a copy after signing an agreement. It's obviously not close to being the same thing as the common meaning of "open source" though.

Hi,

I've found the hMailServer only a few days ago because I'm looking for an alternative for my current mailserver. I'm not yet using it, only testing.

The opensource problem is not very easy to solve. Take for example the Firebird database (you should use this one as a replacement for MySQL ). They have a lot of users, as far as I know the support organization lives on a few large customers. People don't want to pay if they don't have to. But on the other hand, I'd rather pay a small amount of money every year to ensure my mailserver is still supported next year. You could make the software freeware and ask for a yearly fee in return (depending on the number of users served by the mailserver). So people only have to pay if they think they should. Support via the forum is free, direct mail or telephone is only for support contracts.
The hit-by-a-bus scenario is not solved by open source. As long as the project depends on 1 developer it doesn't matter if the source is available. I doubt if anyone will continue the open source project after Martin is gone.
The only problem might be the user community. If someone makes money, other people are usually not very eager to help for free.

Marco.

Hi Martin,

Ok, I understand now . If you are wanting to make that jump from being a part time hMS developer to it being a full time job I can totally understand where you are coming from. I don't think it is being greedy either, I think it is being realistic to yourself and quite possibly being responsible towards your existing users in terms of taking measures to see that development and support is sustainable. Perhaps this is one of the biggest failings of the open source method where developers are not able to make a living without gaining a critical mass of users in their project?

As for the MySQL issue, that is a sticky one. To continue using MySQL with closed source I can't see how you could continue to offer hMS for free with those sort of costs. Could you not build some sort of bridging application to MySQL that is open source? Then hMS wouldn't be connecting directly to MySQL and possibly wouldn't need to move up to a commercial license. I don't know though, I am way out of my depth here so I might just zip my lip . Who knows, but after your last post I certainly have a better understanding of your motivations I think, and all the best too you .

@ Marco

The hit by a bus scenario does work with most of the more popular open source projects. I think hMS is an exception to the rule here as it is unusual to have such a mature and well supported product that has basically been developed by one person.

MySQL's licensing is a royal nightmare. I don't even think the mysql guys know whats in it to be honest. hMail can still use (By use I mean connect to) MySQL but the end user will have to download it and install it themselves. hMail cannot ship with it as it currently does.

MySQL isn't hard to setup and configure but to a new user this may put them off. I think if there are some straightforward step by step instructions on here it shouldn't be that much of an issue.

Maybe support for firebird or other database systems will be added in the future. If V5 has the ability to support plugins which was the original idea then perhaps specific database plugins can be created sort of like a bridge between hmail and the database similar to what you suggested brash. The only issue I can see with that is performance related.

I guess we will have to wait and see

As long as the plugins don't attempt to use DSNs, I'm all for it. Although I'm a bit saddened to see Martin close the source, I understand his decision. I just worry that by discontinuing MySQL bundling (due to licensing issues), he may actually be dwindling his future user base. There's quite a few people who have been helped in these forums with just basic mail server configuration issues. I can't imagine them trying to tackle a MySQL install and configuration on top of that.

True enough prodjtech but then if you look at it from a cold money making perspective it is ideal. If you want specific dedicated support you pay for it. Personally I don't think MySQL is hard to setup especially if your RTFM

I will try to not shoot myself in the foot (Swedish expression for causing yourself problems...) by making the software too complicated. It's hard to remove MySQL without making it more complicated but I'm trying to figure out ways to prevent users from doing the wrong thing. Since bundling PostgresSQL (at least around 15MB) or SQL Server 2005 Express Edition (at least 50MB) will make the installation much bigger, I'm leaning towards distributing different setup's depending on whether or not you already have a database server running but haven't really made up my time yet. I think it feels ugly (from a technical perspective) to bundle a database server with an email server, but I know new users need it this way.

(I've ditched plug-in support for version 5 though. There will be new events and so, but no new framework for adding "big" functionality from outside. The motivation for leaving it out is that I haven't found a way to do it which isn't very complex (time consuming) and the interest in it has been fairly low. Making a generic plug-in framework where you can add for example support for new database types without loosing performance is complicated..)

As Martin sais, you need an out-of-the-box working version so users can evaluate the software without requirements.
For a PHP project which needs Apache and MySQL (and PHP) I created a simple webserver wrapped it in an installer with MySQL and PHP so users only need to press 'install' and that's it. Because the webserver is not open source, I contacted MySQL and asked them about the licensing. They confirmed it was against their license, but as long as I didn't make any money out of it (the software is freeware), they won't do anything about it. I don't like this situation, so I solved the problem by releasing the webserver as freeware and create an installer which bundles the server and MySQL. The script of the installer is available.

Martin,

All I'm saying is if you go down this road, just make sure the documentation is dead-on accurate. But, even that in and of itself can be tricky over time as MySQL may change their installer options, etc. Having written some step-by-step instructions before, I know how time-consuming it can be...and equally frustrating to know that what took you 10 hours to write, has to be re-written because something changes.

Unless...and here's a big exception...you fork the code and allow hMS 4.x to remain open source....rename that project OpenhMailServer. That provides casual users some easier installs (since it can be bundled w/ MySQL) and those who need a mail server for production use, can utilize the evolving hMailServer 5+ code.

Food for thought.

From discussions I have had with Martin my understanding is V 4 and below will remain open source. Only V5 and above with be closed.

For the record it has always been my view that this code for hMs should be closed in its current environment, as basically everything is done by one person here and as far as I am concerned then its only fair that Martin can make some money out of this.

The alternative is simple, at some stage Martin will get sick of supplying his time for free and will stop all development work on this project, the project will then die.

I dont think that the decision to close the sourse is that big a deal as nothing will change for all the hMs users here, as was said above it will still be for free.

Michael

I agree Slug. I would rather closed source and have a maintained codebase than an open source app where the project dies as soon as Martin's giving well dries up...

Pegasus Mail and Mercury 32. Both were closed source. One day developer decided to stop development.

Aida32. Was free, but closed source. Developer sold it to Everest. Refused to acknowledge that it is same product, because then he would have to honor Aida32 licenses. I've moved to other software, which does not provide options that were available in aida32. How much of hardware information in Aida32 database was provided by end users?

Closing the source does not increase amount of maintenance put in final product.

Agreed. (But, I don't think that will be the case with Martin.) I'm just a stubborn ass who prefers open source to closed source. Part of me is just afraid that as much of a PITA as open source licensing is, Martin may find out that having his code closed source may be a bigger PITA...especially in lieu of his #1 ranking on a MS Exchange forum. You know that means Dancing Monkey Boy Steve Ballmer will be gunning for you next, Martin.

Something tells me this discussion could go on forever.

If it turns out that my ideas about this is crap, there's nothing stopping me from changing to open source again. It's not like it's a one-way change.

ok then it's settled.... let's roll out V5!!!

Martin,

You're right, we probably could go on forever. I'll go sit quietly in the corner now....

Is there any way people could get a copy of the v5 alpha?
I feel like its Christmas eave and just cant wait to get my present.

When it's done.. and it's not done yet. Close to done, but still some things left. :-\

Hey Martin:) shall we start countdown?

Uhm......

I've had very limited time available lately so development has been slow. Will hopefully be able to speed things up soon again.

What? Do you actually have a life away from this project.

Martin, I have no idea how you manage to do so much with this project, and you indicate that you have a day job too!!

Wow!

Keep up the great work - at your own pace.

Matt

so let's start the countdown from 100

99

98

Hehe

Dear Martin,
I work for a small company that without HMS probably would have used some outsourcing email provider for managing email.

I would like to explain you why I think HMS should not become closed source.

So far, I never had any need to look inside your code. I'm also not c++ developer, but I tried to contribute as much as I could (posting some scripts, answering posts on the forum, also asking some questions, giving you some feed-backs and so on). I did it because it's my way to contribute (even if it's not much). If have not contributed to any NOT open source project (why should I?).

For me, the source code is just a guaranty that this project MIGHT have a future.
a) If you change your job, have any personal troubles, health troubles, at least there MIGHT be some one whou could continue developing this project.
b) There are too many projects that were Open Source, became freeware and then became Commercial (and also very expencive).
Look at the MySql.

Passing from OpenSource licence to Freeware you are actualy saying that open source doesn't work as a programming model.

I think you should leave the code OPEN and start using some other ways to earn more money like:
1. Selling the email and telephone support.
2. Organising courses
3. Selling implementation priorities (if you pay, your feature will be implemented sooner).
4. Creating an in-box mail servers. (Selling hardware+software in a box, ready to use and configurable by web interface).
5. I'm sure other people might make smart suggestions.

Please don't become one of many other projects that passed pass from Open Source to freeware and then to commercial.
I'm sure that people will be more happy to contribute in any way to an open source project then to a commercial project, and if you ask for some more help, I'm sure you'll get it (as much as everyone is capable).

Sorry for my bad englis and please let me know what do you think.

I disagree, I think Martin is doing the right thing here. Since he is the only person working on hMail he should be entitled not to be ripped off by unscrupulous companies that can and have taken hMail source code and used it for their own gain. I am sure Martin will make provisions and / or open up the source to all if / when he moves on.

I love open source and I support it whole heartedly but I also see that some times open source just doesn't work especially if it's only a single developer. hMail will still be free and as you said you have never had the need to open up the source code so to you this should not make any difference. Of course you have your own opinion and I respect that but I also respect the hard work Martin puts into this mailserver day in day out for free and I for one support his decision to close the source.

standern,

You write that (open) source code is a guarantee that the software might have a future. That doesn't really sound like a guarantee to me. If you take a look at open source project in general, you'll see that a huge amount of them are just ended without anyone picking it up. Having hMailServer as closed source is just as much (or more) a guarantee that it might be continously developed as having it open source. And if you read my previous posts in this thread, you'll see that I will probably give away the source code to companies and people I trust will not redistribute it (escrow are used by a lot of companies for this purpose).

I know that many open source projects are turned into closed source. However, I don't see that as an argument that I should continue to release hMailServer as open source. That's a bit like saying that there are so many companies who are charging their customers, you should start to give me your services for free.

And no, I'm not saying that the open source model doesn't work. It works fine for a lot of projects, specially larger ones wihch are sponsored by companies who invest in them to gain good will. This doesn't work for smaller projects since there's to few users to get good will.

I do understand that I could make money in other ways. I could for instance start to work for Mc donalds, become a sales person, driving taxi, becoming an airplane pilot, and a lot of other things. I'm not sure how this is relevant though. As (I think) I've written before in this thread, I want to spend more time on actually improving the software - not less..

If you feel that you've wasted time by providing support to users who are using an open source product which will turn into closed source, I can perhaps comfort you by informing you that I checked your posts and you've requested help and info over 3 times as many times as you've provided help to others. (Not counting the feature requests you've made)

97

Kaan1983, are your days really that long?

4

How u like me now boy

Heehe I just don't wanna put pressure on you...

Back to 96, relax and code Master

Hehe

Hm, v5 is becoming a closed source product? I understand your reasons to close the source, but you should know open source has his benefits. For the users anyway.

- The future is more sure. I agree with standern. Perhaps this project will be dead one day, but still there might be a lot of users. What if a huge security bug will be found? As closed source project, the product would become worthless. As open source project, someone might be able to fix the bug and spread a solution. Also: There are some open source projects on the net which are quite popular. The original developers stopped, but the fans made a 'jork' so the product continued. (see wikipedia). So a more "secure" future is not just a theory, it's real. Of course this isn't the case with all open source projects...
- It feels more secure. I as user have something like "Hey, this product is open. They don't have anything to hide and 3th party people might have checked the code". So as user you won't suspect hidden spyware or something like that.
- For people who can write own code: They can make own modifications for them-self. Perhaps not interesting for other people, but it might be useful for some individuals.
- People can use your code to learn. I can write PHP, I learned it because I looked and altered existing open source PHP programs. At this moment I can write PHP at a very nice level, I would never be able to do this without the many open source products there are available.

I made the decision to try hMailServer because of the fact is't open. It's just because open source products are having a great reputation in general. When I'm searching for Windows software I get a lot of spyware or shareware results. The word "Open source" is the opposed of that.

For so far my unasked opinion. Of course I still respect your choose of closing the source code. I really do

(3.... )

Now the arguments are starting to repeat themself. I've heard of those arguments several times before and have responded to them just as many times as well.. (Except for the last one. Sure, people can learn to code by reading code, but if your goal is to learn how to write code, there's already millions of lines of code which can be used for this purpose.)

I know open source software has benefits. It's not like I decided this in a few hours.

yeah you are right.
After i submitted the reply I was thinking to edit it again.. welyeah, I was too late for that, you are too fast.

I am all for closed source... and I would not be against a reasonable fee for the software and support. Martin, it is time to get paid!!!! I don't how much money you have taken in over the years from this software but I am sure it does not even come close to compensating you for all your work. When I switched to hMailServer, we were looking at other options that were thousands of dollars a year. I think you should start charging for anyone that uses more than a certain amount of domains and/or accounts. You could create pricing levels based on the amount accounts and/or domains, with an Enterprise version that is unlimited. You could easily charge $1,000 USD one time fee with a $500 yearly support contract (required to receive updates). I know my company would pay. And then you could work on hmail as your full time job.

I don't have 1000$ neither USD nor Canadian hehee..

I belive Martin did a right decision when starting this business.
And I belive he will make right decisions now and in the future

the right decision for him or for you?

I agree with this and I don't have that kind of money to spend either.

I am extremely happy with 4.4 which will remain open source and free so it's not as if Martin has left anyone high and dry.

1. It was not wasting any time. It was trying to do my best (as many users are doing). At the begining, you ask, hoping that the answers you get might be usefull to someone else too.
2. I believe that feature requests are just SUGGESTIONS on how to IMPROVE the product we all use.
3. You haven't probably counted the answers to the polls.

I repeat, I'm not saying I wasted my time because I tried to contribute in the way I could (since I'm not a c++ developer or hMail guru), and I hope to be able to contribute even more in the future.

Anyway, I was just saying my opinion, as anyone does, but I'm sure you'll do the best choise for all of us deciding wheather to release hMS as open source or closed source and I'll continue using it because it's the best product I've found so far.

By the way, I'll confess you how I got to know HMailServer.
Since I had budget=€0 to put up a small mail server (my boss didn't even want an internal email server), I typed "email server open source" inside the Google search engine. HMailServer web site was the firs result. I donloaded it and fell in love with this product.
It's just my opinion (you might disagree but we are just putting together our opinions) that in the future WE might have lower increase of the new users looking for the Open Source products on Internet.

Anyway, Martin, thank you for a great job you've been doing with HMailServer and I hope to see HMailServer V5 very soon and to test it on my server.

hmail v4.4 will still be open source.

I have said it before and I will say it again, I have no problems with the source being closed. I also agree with what Matty wrote above. if you used "x" domains/accounts then you will need the paid version, which may come with some type of support (priority email ?).

I will also say this, no one here is being disadvantaged by the closure of V5 as it is still FREE. I have been here now for about 2 years (if not longer) and I don't remember anyone submitting code, so the argument about people helping/checking the code is not relevant in this instance.

If this project was squirrelmail or firefox etc were it had 10 or 20 developers then I would agree that closing the source would not be in the projects best interests. But hMs has one developer, and he can do with it as he see's fit.

Michael

I've been avoiding comment, but I agree with martin's decision. Closing the source is in his best interest to protect his intellectual property. He is still keeping it free, with all the features available here I wouldn't mind being charged for this product (not that I have the money to at this point).

Slug makes a great point, there is a single developer. I have been here for probably a year, I did download the source and look at it, but my programming skills were not good enough to contribute to this project.