[Estimated Reading Time: 2 minutes]

Marco Cantu just blogged about the price of Delphi and his conclusion that it is reasonably and comparably priced. Whilst it might be encouraging that he is looking at these things in his new role, his methodology and reasoning leaves a lot to be desired.

Someone in the comments to his post made the point that it was Borland who hiked the price to current enterprise levels. That was when the complaints started, and they haven’t quietened since. So it is not surprising that if you look around now that you will find that Delphi is priced comparably with it’s competitors – of course it is, that is the complaint in the first place !!

The problem being that those other competitors have other things going for them. Looking only at price or the licensing model misses the point entirely.

If Delphi is priced only as well as it’s competition then it simply cannot beat that competition on price but must compete in other ways.

It cannot compete in terms of “career prospects” – the number of Delphi jobs continues it’s downward trend – and sadly it no longer competes on “customer/user experience” (see point on innovation later).

It never has been able to compete on the basis of “industry standards/acceptance” and at the current price, it has no hope of competing in the hobbyist space – the space from which enterprise validity often springs as it builds the foundations on which enterprises rely (a pool of available and competent skills) in the absence of any formal support (Delphi is not taught in schools, and increasingly neither even is Pascal, so these skills need to be nurtured elsewhere and elsehow).

But – some will cry – Delphi competes with innovation! Just look at FireMonkey and the multi-platform support!

Being innovative can help, but only if that innovation is something that can be relied upon. FireMonkey and FM2 are still not up to scratch as far as reliable code quality is concerned, and introducing features and then yanking them unceremoniously as was done with the iOS “support” and was attempted with the Professional client/server/n-tier EULA changes is not a good way to establish trust among your customers as far as product/relationship reliability goes.

So, given that Delphi falls sadly short in all these areas, price is the only significant lever left to pull, and selectively pulling out a narrowly focussed price comparison to justify not doing so (and indeed in order to defend the most recent price hike) is simply further evidence of the 1990’s “enterprise” mindset that dominates at Embarcadero to the extent that it has already pervaded (or was already present in) the mind of the newly appointed Product Manager.

36 thoughts on “It’s Not All About Money”

  1. He and borcadero are seriously out of touch.
    I won’t be purchasing Delphi anymore, it’s simply
    To expensive for small shops and hobbiests/enthusiasts
    When you can do cross platform and then some with Lazarus.

  2. Can’t agree more. The real issue is which public the product is targeted. As mentioned, current prices are at enterprise level, so enterprises are the target destination.

    Being an all-purpose & minoritary product, its natural “echo system” (where it has more chances to grow) is not enterprises but students/hobbyst/self employed/small companies which simply can’t afford it.

    Obviously innovation helps (a lot), nowadays most innovation comes from the opensource arena (ie. students/hobbyst/…)…

  3. I disagree. It is all about the money, at least for EMBT. As a business they’re understandably very concerned about generating revenue, but they’re mistaken about how to do so IMHO. It’s not by bundling more products and features in the box. I would be all for removing many existing features and selling the core product for less – a la carte. I believe they need to focus their efforts – do less, and do it better than anyone else.

    The best way to generate revenue is to produce a best of breed product, with compelling feature(s) to attract new developers and keep existng ones. That compelling feature(s) could be FireMonkey and XPlatform (including mobile development) if it’s done well. If market adoption (reflected by commercial software releases based on the technology) is any indication, the first FM release with iOS support was not well received. EMBT probably only has one more opportunity to prove they can deliver a viable IDE and UI framework ready for commercial devellopment.

    1. Um, you disagree ? Are you sure ? I think we’re both saying that Embarcadero shouldn’t assess whether or not they are fairly priced by looking solely at the $$’s on the ticket. Embarcadero’s response to complaints that they are too expensive is focussed solely on the money: “Our products are priced comparably with others. Suck it up.”.

      We both seem to be saying “It’s not that simple”. That it is not just about the money. No ? 😉

      On your other points, I disagree on your analysis of how to best generate revenue. Simply creating something (no matter how good) is not enough. You also have to be able to sell it. That is, increase sales volumes, not just the value of sales. And the quality or “excitingness” of dev tools isn’t enough these days to win sales.

      I also disagree with your hopes for FireMonkey – with FireMonkey they have painted Delphi into an even smaller niche than it already occupied. The VCL was already a minority interest framework, and now they have created another framework with even narrower appeal. A framework that has one big feature – multi-platform – which unfortunately in no way offsets it’s many, many shortcomings.

      Even if someone is excited by the potential of FireMonkey, they are faced with investing a lot of time and effort into learning a language and framework which contributes nothing to their knowledge of the platforms which they may be developing for, so they are taking a huge risk in adopting the tools from the one vendor that supports FireMonkey, and severely restricting their appeal to the labour market should things not pan out and they find themselves needing to find alternative (or supplementary) employment.

      Of course, this doesn’t matter if you can use FireMonkey to make the next “Angry Birds”, but if that’s your approach then you’re probably better off taking the money you would spend on Delphi etc, and buying lottery tickets instead. 😀

      1. About the idea of Delphi a la carte, Firemonkey will not be the focus UI framework, because with a cheaper core product you can buy, develop o download a different framework (Gt, gtk, etc), just like dbexpress and datasnap, there´s other frameworks, like Zeus db objects, Aurelius, etc. So what i see is more migration to other pascal alternatives or even other languages, and of course pirate enterprise delphi. please correct me, but in the beggining turbo pascal was an affordable product with similar power and functionality, nowadays is as Marco says similar in price, but with less power.

        greetings

      2. > with FireMonkey they have painted Delphi into an even smaller niche
        >than it already occupied.

        I touched on this in a reply to Marco’s post about 2013 being the year of Delphi and that Delphi was “coming back” – a fascinating term since no one at Embarcadero will ever admit that Delphi’s glory days are long behind it and continue to claim that nearly 1/5 of developers around the world use Delphi (they cite 3mil users and 16 mil developers worldwide, an absurd and easily debunked claim).

        The mobile revolution or “post PC era” began about five years ago – further back if you count the rise of web apps. Delphi users who were interested in entering this era and creating mobile apps have already chosen another development environment. This is the same with 64bit and unicode… surely any C++ Builder user who needed 64bit didn’t wait around until 2012. This means that the Delphi users who are left aren’t that interested in mobile development. Meanwhile, VCL is obviously mothballed – they don’t have the manpower to develop two frameworks simultaneously, just as Swindell said of the time at Borland about Kylix and the VCL. Given the layoffs they presumably have less people now. If many Delphi users have been using the same language and framework for 15 years or more, how easy is it going to be to get them to switch to Firemonkey? Delphi still exists _because_ its remaining base have no desire to switch frameworks and don’t care about anything except Win32. Its one strength works against it in this scenario.

        Factor in the unfinished state of Firemonkey and the abominable user experience for the iOS toolchain (while it remained in the product) and I don’t see who in the world is going to avoid the free or almost free official tools available today for mobile development to use a language they don’t know to develop non-native apps. Delphi users don’t appear to have a pressing need for cross-platform, either, so who the heck “Develop your mobile AND desktop apps in Delphi!” appeals to is beyond me. If people want all-in-one solutions they can turn to products like Mono (.Net) that will let them use the language(s) they really are using now to develop mobile apps. They’ll also have official Qt support on Android and iOS this year (unofficial ports already exist and it’s a stable, mature framework that even has an LGPL license available).

        I think being late to this game is going to be worse than not showing up at all. Anyone interested in this has already left and it’s going to shift development away from the one thing people still like about Delphi. It’s Delphi.Net all over again.

        Marco sees this as “FUD”. I think the fear, uncertainty and doubt is warranted.

        1. Execllent analysis, and I agree whole heartedly. FireMonkey could have been the source of ressurection of Delphi but they entered the game late, and took on too much (a UI framework and XPlatform compilers), or it’s demise. Based on Chris Rolliston’s assessment of FM2, the late delivery of new compilers, staff losses, price increases and the other points you made, I tend to side with the latter.

          If EMBT embraced the community (IOW leveraged open source as well as listening to their customers) then perhaps they would have sufficient resources to fix the things that have fallen into the black hole (aka QC), and make the enhancements that need to be made for Delphi to gain market share.

    2. EMB are simply seeking customers who are willing ot pay their price.I think they are aware of what they ship. There is always the one or other amazing thing – FM2 Designer – respect.

      Assuming everything works fine. I think the just normal guy sees this video

      http://www.maybach-manufaktur.com/

      and will think – a car is a cool thing – need one. After taking a look at the price label the guy will come to the conclusion that he will very likely need another vehicle first in order to be in the position to afford this specific car one day. The comparison to Lamborghini, Ferrari, Porsche and Jaguar will not help the young guy riding a bike.

  4. For commerical users I think it should be looked at as a balance of quality vs price. The equation is different for hobbyists, students, charities etc. But as a commericial user if the quality and funcionality gives me a sufficient return on investment then the price is not an issue.

    Having made a respectable living from Delphi for over 15 years I don’t have a problem with the price tag (forgetting about exchange rate conversion of course). I do have major issues with their product quality, scatter gun approach to chasing new features for marketing purposes and general arrogance and neglect of the user base. The hubris displayed by certain high profile Emb employee(s) over the last few years is also quite staggering.

    But the bottom line for me is that they can double the Pro licence fee for all I care. In return I want a solid product, bugs fixed in current and recently “discarded” versions, no punative license changes, respect for the community and acknowledgement of errors and mis-steps where they are made.

  5. Well, Delphi has one big advantage: Object pascal!
    After 15 years of Delphi programming I’m finally forcing my self to learn C++.
    It’s horrible; Case statement fallthroughs, awkward terminology (why use static instead of class var?) , unpteen different constructors…

    But C++ has better floating point performance (with Intel compilers) and OpenMPI support and CUDA support. And most importantly 90% of programs I see are in C++ (and the rest in Fortran).

    (I apologize if this little rant is a bit of topic)

    1. Object Pascal is great, don’t get it wrong, but the warranties you want are in other languages, like C#. If you develop in Ruby, you will see how weird feels to write either C# or Delphi again (of course, when you can use it, as Ruby is incredibly slow, and has no good visual designer).

      Also both Java and C# are more popular than Delphi can be, and if picked right, they give most of what Delphi strides too for a very low cost tooling (by what Delphi strides for are: Visual Components with Designer, great debugging experience, components, native performance, some even offer static code analysis). Some things can be added to Delphi to address some issues I’ve said before (like Castalia static code checker which is not necessarily bad tool, but is a bit off compared with the free IntelliJ IDEA) but this will increase the difference (in price/features) from the tools I’ve mentioned before.

      And if you think about “native experience” and you think that the VM gives a hit too big when it JIT’s, there are static compilers (for Java is Excelsior JET, and for .Net the default added NGen) that provides the same short startup time of Delphi (and most of times will give similar or even better performance.

      At last, both Java and .Net bring more languages (sometimes better suited for their job than the default’s platform one) like Kotlin, Scala, Xtend, Boo, F# and both platforms offer a way at least to interact with C (in case you need to execute a loop with OpenCL/CUDA or you can afford Intel Compiler).

    2. Ajasja, you’re correct about the complicatedness of C++. However, that’s another older language. I assure you if you try a more modern language like python, often dubbed “executable pseudocode”, which has made readability one of its core values, and which only has 20 reserved words (simple language wedded to a huge standard library) you’ll never want to go back. I call it “the new Pascal” myself.

      You can actually find books for it, read about it in magazines, it’s taught in schools, there are videos online that aren’t just Nick Hodges screen captures or Coderage videos 🙂 (pyvideo.org has at least 1328 videos indexed!), there are free online courses from places like MIT Courseware and Coursera that use or teach it, there are over 26000 (!!!) open source packages to do anything you can imagine indexed at their website, it’s dynamic and interpreted (but you can also use Cython, which is 99% python plus the ability to add static variable types which converts the code to C++ then compiles it!), there are multiple IDEs and GUI toolkits available, it runs on anything you can imagine, from Win, OS X, Linux to Android and embedded devices, it’s open, the community is friendly, the language is driven by the users (users submit and vote on changes and additions), and it’s all free. 🙂 There are also many major web frameworks and CMS systems based on it like Plone, Zope, Django, etc. It’s really amazing. It can also interface with C, C++, Fortran, R, Octave, Matlab, .Net, Java, etc. There are also versions that run IN .Net/Mono (IronPython) and Java (Jython). What more could you want? Free books? There’s several of those available online as well. 🙂

      Hello World is just

      print(“Hello World!”)

      Have a string with a name you want to split into two variables, first and last? If you have the right version of Delphi you could use the new split record helper and put the names into an array of string and then move them to the variables. Otherwise, you could create a TStringList object, put the string in, split it there, and then copy the two names out to the variables. Old school Pascal you’d probably do this:

      space := Pos(‘ ‘, name);
      first := Copy(name, 1, space – 1);
      last := Copy(name, space + 1, Length(name) – space);

      In python it’s just

      first, last = name.split(‘ ‘)

      🙂

      If that looks prettier to you than C++, check it out. Like everything that’s not Delphi, it’s free to try. 😉

  6. Sure the price tag is too high, but we do know nothing about how the company is doing in financial terms. Say, you run a company that for historical reasons has zero new user adoption rate and a number of sure-paying customers. You can leave all things where they are and experience slow decline and inevitable death. Or you can try to do something, with all your actions leaving to the growth of expenses and you are forced to raise the price tag to keep yourself afloat. Otherwise you go bankrupt. I suppose the latter will greatly disappoint all of us 🙂

  7. XE was the last version of RAD Studio I purchased, having suffered through so many, many releases that fixed past bugs only to create new ones so I would continue to spend money. Never again. I’ve stepped off the treadmill and stepped onto the VS2012 / Oxygene path and am happy to have done so. I know that VS2012 will continue to receive fixes long after the next version is out, and I can mix and match future technology. Having been with Borland since Turbo Pascal 1.0 (and having written Qmodem with it), I am not looking back.

  8. Well, Lazarus/FPC also have a very spot on implementation of Object Pascal. I say let them price themselves right out of the market.
    Start using Lazarus people, and if you have the know how contribute to make it better, which you cannot do with Delphi.

    I remember back around 1999 I could upgrade my Pro License for under 200 bucks and it had the no nonsense license agreement. Even as a poor student I could afford that, but it’s way to expensive now and the end user license agreement is a freaking joke. Way to go Borcadero.

    Maybe if all of you stopped whining so much and took the leap to use Lazarus Borcadero might see the light.

  9. After going through Delphi 2009, 2010, and then XE2, and not getting up to speed on any of them, I am not in the frame of mind to splurge for XE3. XE2 has been so buggy, in regards to the paths constantly corrupting from component installs, that I have reverted to using only Delphi 7. I don’t see a future where I have to upgrade every year or less for features I do not use on a platform that is not maintained.

    It looks like my future will be in M$ products, since the money output is about the same anyway. At least there are more jobs available. It is a shame that the money comes from people who are enamored of a name brand though.

  10. I like EMBs Delphi efforts and think the price is good. I, of course, always want to see more features. Visual development features to be exact. But, it takes time.

  11. If Delphi goes legs up at least we won’t have to settle anymore. Between the Free Pascal compiler and RemObject’s compilers the community is doubly reenforced by both efforts.

    I would go as far to say that even without a Delphi to speak of Object Pascal will live on even more so than way back in (old) Borland’s heyday.

  12. Marco’s article is about a controversial topic, I think it’s here to make people discuss. Maybe about getting information and preparing arguments …

    We would not discuss about this topic if the quality shipped fits and requirements are covered completely. There can be no compromise in terms of quality/correctness. The latter is a matter of taste, everyone will have to decide whether the feature set does match his/her expectations and demands or anything else fits better.

    The VS comparison. It about comparing apples and oranges. There would be no demand for something like Delphi if VS allowed you to work satisfactory. Using MS products is never wrong and buy from IBM is very much the same. Microsoft sells solutions for problems they created on their own, so does IBM and SAP … VS is not competition as MS is no competition to EMB. EMB can move Delphi away from where VS settled down but no one can hinder them from maintaining the existing.

    What is important to distinguish, considering opinions based on propaganda. All big vendors use propaganda called public relations and finally manipulating peoples thinking especially the decision makers thoughts. Toady many companies independent if big or small vendor simply sell via modifying your brain. From this perspective a PL/SQL Developer, SQL Detective and PhpED (mentioning a few) are very likely a good choice – they simply convince with the quality of their products. The niche has an advantage – I doubt Delphi is a niche product. Delphi is an IDE that tries to address the demands of many niche businesses as a development technology – achieves this goal via extensibility. VS is similar. VB.

    In the 90s, Jolyon you are totally correct, the thinking was a different one. Oracle never bothered with tools or simply purchased the modeling tool and an IDE that never fit together (Oracle Designer and OracleForms are 2 total different products developed by different ‘companies’ – organizational units …). So it was not unusual to use third-party development environments. The thinking is still 90s – Windows in the 90s.

    This changed … and with this the way development is organized and the way vendors make their money (store) + cloud. If the way to make money is donkey f**k they will develop a technology that allows to effectively harvest donkey f**k.

    I don’t use technologies or run-times that expect a certificate. Such a restriction only frees the big vendors from any responsibility but taking 30% for just the name. Never. Feed the rich … ?

    The point is not the price, quality first maybe afterwards it makes sense to think about extending the customer base.

  13. Hello yes Lazarus for the win 🙂
    Lazarus can totally take care of business.

  14. Lazarus is a top-notch implementation of Object Pascal and a worthy successor to the Delphi crown, I’ve been using it to create Windows 7 and Ubuntu programs for the past few months and it’s refreshing to see a lot of bugs I still see in Delphi XE2 (and I have no chance of getting fixed without a $500 upgrade) are no longer there.

    I Use C a lot for my embedded development but its always good to go back to the warmth of Pascal for desktop work 🙂

  15. Your points resonate well with me Jolyon. My gut feeling is that the Delphi ecosystem is not where innovation happens these days. I would not argue that there is much wrong in Delphis pricing per se, but that value/quality you get for the money is absolutely not competitive these days. Sadly enough this trend seems to continue too.

  16. The innovation is going on in the Lazarus community now
    Without a doubt. I really encourage Delphi users to start
    Looking at Lazarus now as it really is a better choice in my
    Opinion. I see a lot of people asking for a Delphi starter
    Edition and it’s like da people you have that and more
    With Lazarus. The more users Lazarus gets the more third
    Party component devs will take notice.
    I can now spend more on components with Lazarus as I don’t
    Have the huge yearly upgrade expense.

  17. I too would love to see more of an a la carte pricing and delivery structure. Maybe the pending “mobile add-on” is a sign of things to come? I think it was really stupid to introduce mobile support, then take it out and say “now you’re going to have to pay for it unless you’ve got one of the more expensive products with SA”. WTF?

    I have NEVER worked at a place that EVER bought anything OTHER than plain vanilla Developer licenses for their folks to use. Some of them even have corporate policies against buying third-party components, unless there’s an extremely compelling reason to do so. So I have a really hard time understanding where their market is for the higher-priced SKUs. Who buys the Architect Edition for their entire development team, especially when the Developer Edition has everything they need?

    Why isn’t DataSnap available separately, rather than as part of an entire product upgrade? If the Mobile part will be released as an “add-on”, then maybe DataSnap may follow?

    1. Datasnap was separated once. Datasnap was called Midas before. The initial price of Midas was about 15k EUR about 20k USD iirc – considering inflation 50k USD (in order to compare). So it was no big surprise that many MIDAS replacements have been implemented. Afterwards Datasnap was sold seperately at 1000 USD …10-12 years ago. These days Borland offered the Visgenic ORB on one hand and the Datasnap on the other. The latter was aiming was focusing on integration at the data level. Then something happened no one expected.

      After Microsoft built the MTS ‘on top’ of the SQL Server 6.0 or 6.5 that was not in the position to handle more than about 20 users concurrently the former enterprise software architects, who failed on Smalltalk the first time said, ‘Cool. let us put something on top of the databases too then we have our Smalltalk back in Java’ and they tried the same thing that did not work before again and again. CORBA and C++ lead to executables that ate up all the computers memory.

      Inspired by design patterns and only knowing the proxy design pattern when it came to distributed computing inspired architects decided for a crap piece of software called Weblogic Java Application Server because it run on Sun. Finally it turned out that the Proxy did not scale the way they expected. Oracle + Sun and Y2K. Peoplesoft … these days. These jewels turned out to become expensive because putting things on top of other things is the usual response – hiding the sins from the past.

      The parallel movement was Java + CORBA leading XML + Web Services finally evolved from this.

      Why not CORBA … the idea these days was another one. Have a network with computers have demons running as Singleton (context of the machine) and those provide services. For example a service collecting data about a city. Every city has it’s own server but the interface is the same everywhere. When you wanted to know the number of inhabitants ask the appropriate service at the server in New York and if you want to combine it with San Francisco ask 2 servers. This was accepted thinking of the object oriented people these days. No fun. Those visionaries deeply influenced the way applications are designed today.

      The former Smalltalk illusionists didn’t stop. They wanted to put into practice their vision of distributed computing on object oriented databases + Smalltalk. Today the 90’s vision of Versant database that never made it in a roundabout way into the Telerik ORM is one example. The difference to past is simply that the collections are not stored into the database directly and had deep impact on the database design and RTTI instead of pre-compilation steps. Since we have XML we can store objects so that different technologies can read the data from the ORM.

      A big business in IT is usually based on a false response to a problem that has not been understood or simply didn’t exist. I think one good response to the problem was simply using IIOP and integrate it into the Linux ‘desktop’. 2 decades of trying to build the illusion of 3tier … every 2tier approach works better, but no one can make billions of dollars with such an approach when talking about tool and infrastructure vendors.

      The cloud is about getting rid of these Appservers too. They are too heavyweight.

      The data-snap was never wrong. The rest stuff has been built on top of a concept that is sufficient. Coming back to the beginning Borland always wanted to bind the ‘CAL’ to the Client Server version of Delphi. This did not change and all these strange decisions never changed. So EMB cannot be ‘the one’ who does the decision – maybe someone at EMB that continues the old game Borland played.

      Before the data-snap can be unbundeld a redesign is required. The datasnap integrated into a box hosed does make sense, maybe together with the interbase. EMB obviously decided that it would be our freedom to do so … they miss an opportunity if they stay on the current way. This is the problem with Delphi in general. You have toolkit that would allow to construct everything but EMB adopts Delphi in a way it servers their sales best – in this case you are right you simply need the ‘pure’ Delphi.

  18. Sometimes people don’t remember how poor was TP. It couldn’t make .EXEs until release 4. It didn’t have a debugger until version 5. It didn’t have a GUI lib until version 6. It couldn’t use DPMS until version 7, and only if you bought the more expensive Borland Pascal 7. It didn’t have visual designer when VB already had them. Sometimes people look at the past with pink glasses. TP had a good price/feature ratio, price was low but often features were too. Delphi was designed as a more professional tool than TP was, and the price increased – and features too. The problem now is that compared to competition price/features and price/quality ratios are not good.

  19. Sry – Today the 90′s vision of Versant database that never made it in practice, finally made it in a roundabout way into the Telerik ORM is one example.

  20. I think borcadero is beyond help now. I encourage
    You all time take a look at Lazarus its a viable alternative

  21. > Whilst it might be encouraging that he is looking at these things in his
    >new role, his methodology and reasoning leaves a lot to be desired.

    I agree, and one major flaw in the method/reasoning in my opinion was this decision:

    >Of course, there are open source development environments… which are
    >free and make the comparison unfair. In most cases… they were
    >originally build by a company who focused on a different business…. How
    >can we compare those projects with an IDE build and sold by a company
    >which is into that specific business (development tools) and not a different
    >ones (like operating systems). Honestly, this isn’t easy, so I’m not trying to
    >offer a complete comparison. I don’t think one is possible or could be fair
    >and honest.

    Realistically, “fair and honest” doesn’t play a factor in cost comparisons. A consumer doesn’t say to themselves, “I know this Windows handset maker needs to pay a $50 license fee to Microsoft for the OS but this Android handset maker gets the OS for free, so I’m going to mentally add $50 onto the Android handset when doing my comparison”.

    The reality is that the landscape for development tools has drastically changed since the ’90s – for the most part, they’ve become a commodity. There’s also a heavy favoring of standardized and/or open source languages, leaving Embarcadero’s corporate, proprietary, “de facto” language standard approach something of an anachronism in 2013. The consumer didn’t choose EMBT’s business model and isn’t responsible for it. To draw the conclusion he did while simply ignoring the “elephant in the room” of open, free language development tools makes the conclusion somewhat useless. He’s attempting to decide whether the price is in comparison to a specific, limited subset of (as others have pointed out, often more featureful or higher quality) commercial products, while the real question is how the price stands in comparison to *the market*, in which the commodity tools play an enormous part. In essence he’s addressing whether this is a fair margin for EMBT rather than is this a price that the market will bear. It’s quite possible that EMBT’s price is reasonable but the market simply won’t bear it given the other, free tools available.

    It gets even worse when one factors in an interview EMBT’s CEO gave to The Register. In it, Mr. Williams is queried about what specifically Delphi offers that C# doesn’t (a question I can’t get anyone at EMBT to give me a straight answer to myself). Williams states that C# “is .Net” thus it “isn’t a competitor”!!! This is like saying that the Toyota Prius isn’t a competitor of other automakers because it’s a hybrid and not a conventional gas engine. Delphi might not compete with C#, but rest assured C# is competing with Delphi! The interviewer then shifts to MS VSC++ and Williams dismisses it as “not their focus” (I’m sure C++ Builder customers must love that). So between the CEO waving away C#, C++ (and presumably Java since that targets a VM) and Marco using business models to discount comparisons with open source tools like PHP, Javascript, python, etc., that apparently leaves Delphi with no competition whatsoever!

    One gets this feeling when examining the marketing materials on the website. Nowhere is there a feature-for-feature comparison with the Big Guns of modern programming (C#, C++, Java) or a case made for Delphi’s superiority. It seems that despite all their bluster, they might really believe that their only competition is their previous version and know that they’re just selling to existing Delphi users. On the other hand, Marco Cantu wrote that “I’m convinced that in the business world Python has a fraction of the Delphi influence” (despite the upcoming PyCon being sponsored by Microsoft, Google, HP, Rackspace, Netflix, Disney Animation Studios, Lucasfilm, Red Hat, Ebay, Facebook, Amazon, Canonical, Twitter, Dropbox, Oracle, American Greetings, Reddit, O’Reilly, Hulu, Github, CPanel, and many smaller firms with over 2300 attending in 2012, while CodeRage is a virtual-only event sponsored only by half a dozen Delphi tool vendors). In some ways they still don’t understand (or wish to accept) the market Delphi has built for itself. This analysis did nothing to change that.

    1. Are there even major physical conferences for Delphi anymore? I found one for Delphi Live (in America) but the web page hasn’t been updated since 2011 so apparently there was none last year or planned for this year. I also can’t find any counts of attendees for any of them. I did see that the price for attendance was about the price of a new copy of Delphi Pro too. 🙁 Any photos I seem to find of these conferences or of an event Marco is at don’t look like there’s more than 30 people in a room.

      On the other hand… there’s a Lazarus conference? Awesome!

      1. The dutch pascal/Delphi user group has an annual Lazarus day in Utrecht (typically in june, though this year it was sept 1st)

        Usually 8-10 FPC and Lazarus devels are there too then.

        1. That’s not bad at all considering I found a page that claimed the Italy Delphi conference was the largest in Europe in terms of speakers and such and in 2012 it had 15 speakers and 70 attendees (and 32 prizes!). FPC/Lazarus is definitely gaining on Delphi.

  22. It sure is, I love it and I got to it first over Delphi since I can
    Write Linux service apps.

Comments are closed.