Since the beginning, still during the project's conception, my intention was to deliver a quality open source product. So back in 2012, when pgModeler gained the world, its user base started to grow faster than I could imagine and many feature requests and bug reports flooded the project's issues page at Github. I was really happy with that feedback since there was some interest in my work. All that activity around the project was really new to me so in order to give users a quick response I was needing to upgrade my workstation in some point in that year. Prior to 2012, specifically between 2006 and 2009 pgModeler (called __pgsqlDBM__ in that days) was developed in an old frankeinstein desktop which I honestely don't recall its hardware configuration but I'm pretty sure that it was not suitable for software development. The work in that machine was painful since its sluggish performance was difficulting the creative process due to the slow compile time and some other issues. Thus, still in 2009, I decided to buy a budget Dell Inspiron 1545. Well, that was a great improvement and made me work hard to release the first pgModeler three years later. For the years to come until 2012, pgModeler was distributed __free of charge__ in pre-compiled binaries only for Linux 32 & 64 bits and Windows 32 bits. But that situation doesn't took too long because still in same year some Mac OS X users started to mail me asking for a port of pgModeler to their beloved OS. I have accepted the challenge and I was willing to give them the deserved support but I needed to face the reality of many open source projects: __the lack of financial support__. Despite the fact that I was alone in this project and not a single penny was request I was planning to buy myself a dedicated Mac machine but here in Brazil due to the abusive taxes and the constant economics crisis buying such equipment is for few! So I had to improvise by taking borrowed some friend's Mac machines or even using virtual machines to run a Mac OS X so I could barely do my work on that system. After some months, in February 2013, the first port of pgModeler was released for Mac OS X and I need to say that was a real disaster... _as I was expecting!_ No dedicated machine nor enough time to work on that OS led me to a big frustration as well the interested users. So many negative feedbacks that I almost dropped the Mac OS X port. Instead of giving up, I decided that was time to get a proper machine but again due to the abscence of financial support I needed to charge for the binaries later in 2013. But that wasn't plain simple as that, first I tried several contacts with some Enterprises to patron the project but no answer was given, unfortunately. Thus, after many complaints, little by little, the users understood the reasons why I needed to distribute paid versions of the project. Obviously its essence was not left behind: be a quality open source software. This way, with the help from donations, purchases and, of course, my paycheck from my "real" job I did my first real investment to the project: buy a Dell XPS L502x notebook. That was a great leap in the performance and I could respond more quickly to users' requests. With the help of that machine pgModeler could be released in Windows 64 bits because, until then, the framework I've used to build it was only distributed in 32 bits and I had to compile it from scratch on the 64 bits environment! But, the main problem still persisted, the lack of a decent Mac OS X environment... I could run the OS with a better performance in a virtual machine and make use of the machines from my friends but that was not the ideal scenario. I've used that great XPS notebook until last month (March, 2017) when I needed to replace it by a Dell XPS 8900 desktop after a hardware fault! Well, that new machine is really marvelous and I can compile all five releases (Linux 32/64, Windows 32/64 and macOS) in few hours instead of half a day as was done with the prior computer. But the macOS issues were still there... and I still couldn't afford such expensive machine like Mac. Thanks to all the good people who believed in pgModeler's pontential and helped it with some money that new XPS desktop was bought using last year's donations and purchases. I really would like to acquire a Mac machine too but I would need at least __twice the money__ to get one. That's why I have created this campaign with a straight goal: to reunite resources more quickly so we can have a good machine and give pgModeler a decent performance on macOS.__Why charge for the binaries?__
In short __to keep alive the project!__ I believe that ethically, the source code needs to be freely available to everyone who would use the software. But pre-compiled binaries are a convenience to save users time. They are really a "support service" that I have been providing to make it easier for others to use the software. It's a compromise, but not a harsh one. And for me, it is ethically correct while being financially practical. I value Free Open Source as a philosophy and for many practical reasons. It was always my goal with pgModeler to deliver a quality open source project. For me the "Free" in FOSS is for "Freedom". Therefore pgModeler is and will always be __Free Open Source Software__. This is the reason it is licensed under GPLv3. But while I consider it of highest importance to provide the source code to the community without any barriers, providing the most up to date binary for each platform is extra work, and is really for convenience of the end users so they can have it working faster, and know that the binary came from the original developer. To be very clear once more, the source code is totally free. This is not freemium software. There is no "Pro" version. It's zero cost, full functionality software. You get all the features whether you pay for the convenience of a binary, or take the codebase for free and compile it for your platform of choice. And I hope many people do compile from source and join the community to make the project even better. I try to provide clear and easy to follow directions for compiling, but if you have questions, please do ask for help so I can improve the docs. FOSS projects have always struggled financially with the challenge of giving the source code with no restrictions on what can be done with it. Because there are real costs to making it, and very few ways to receive payment for the work. So the option I've chosen is to allow everyone to get the source without restrictions, for free, and to charge a small amount for the convenience of a ready to use application. I am essentially charging for the service of maintaining pre-compiled versions.__No corporate conflict of interest!__
This project has no company paying for the time or equipment. And unless I can find corporate sponsorship that doesn't have any strings attached, I hope to keep it that way. Too many projects get bought out by companies and then killed. Or the companies pay for the project just to hire the primary developer and then give them so much other work that they never spend time on the project anymore. I don't want that to happen to pgModeler. But with a family, the time spent in the project is time away from my wife and kids! And living in a country with low wages, and high taxes on imports, it's quite a financial sacrifice to get the equipment necessary to make development easier, and to release code faster. So, while I never want to restrict access to the project based on ability to pay, any income it does provide to me will enable me to afford more development time on it. And right now, it would also enable me to multiply my efficiency by getting some new equipment.__Is the macOS version that bad?__
Short answer: __YES__! Long answer? There's so many problems in macOS port that's is really hard to start talking about them in a simple way. Well, there are countless problems in macOS, from clunky interface and crashes to unexpected bugs that happen only in that system and I don't even have the real chance to track and fix them as it should be. One of the most problematic situation is when a framework update breaks things in pgModeler and I have to choose between two options: downgrade the framework used to build the code and not take advantages of optimizations and new features; or keep the new framework and patch the code where it's needed. The first case is the worse choice since we'll never get improvements of the core code running in our software. There were cases in the past that the Linux and Windows releases of pgModeler were running a Qt version two numbers ahead compared to the macOS release due to the broken code that I couldn't fix properly. After too much work I could match the Qt framework version used on the three systems but specific problems still lurks the macOS port. The second case mentioned, that is to adopt a newer framework release but doing the needed fixes, is what I personally prefer and is what I'm currently struggling to achieve but the lack of a Mac machine makes things quite difficult! I really don't like or even think it is fair to have two good performing releases (Linux and Windows) and left a poorly maitained one for macOS users! The fact is, in the day this campaign was launched, the macOS user base was the second one or __24.1%__ of the total audience losing only for the Windows 64's user base which was __36.8%__ of the overall number. The least ones were Linux 64 (__20.5%__), Windows 32 (__10.9%__) and Linux 32 (__7.7%__). These numbers only emphasizes the importance that we should give to macOS system and its users.__Why a Mac mini?__
Right now, my biggest need is an efficient development system for releasing for macOS. I've determined that the best option for the money is a Mac mini. It's powerful enough to do everything I'll need for a number of years to come, and the cost is low because I don't need it to be a laptop or a powerful desktop like the Mac Pro. I just need a simple little workhorse that runs macOS natively and doesn't depends on anything else besides an energy supply! I know that there are online services that provide macOS machines at a low price but since they depend on internet access, which is really inneficient here in Brazil, own a physical machine is the best option because I can even work offline if I need to! :)__But what I get if I decide to help?__ _My eternal gratitude in the first place!_ Followed by an even more improved product and finally, and not less important, an __access key with an extended validity__ depending upon only on the amount you donate! ;) If you don't know yet, the access to pgModeler's pre-compiled binaries are granted only for those who have paid a small charge in the download page. In that case, the user receives an access key with a limited validity that may vary from 6 to 24 months. If you decide to help us you'll receive a key with full access to the pre-compiled packages for __all supported platforms__ (eBook is not included) and a longer validity which is calculated according to the amount of your donation: _for each 10.00 USD or 30.00 BRL you'll get 3 months of access key validity. **Donations less than 5.00 USD or 15.00 BRL will not be rewarded with an gift access key.**_ Let's say, for instance, if your donation is __10.00 USD__ you'll get an access key with __three months__ of validity which can access all releases and packages for Linux 32/64, Windows 32/64 and, of course, macOS while it is not expired. Donating a greater value like __40.00 USD__ you'll be granted with an __one year__ access key, and so on. Once expired the access key there is no automatic renewal and the access to downloads will be blocked. __Note that all rules in the Terms of Purchase & Download (en_US, pt_BR) are applied to the generated access keys.__ Isn't that great? So, what are you waiting for? Let's help pgModeler! :D If you have any question about this campaign feel free to mail me via __Support__ menu at the top and don't forget to share this page so we can reach the goal faster!