#CodeRageX: Idera
#CodeRageX: Idera
David Intersimone described the Idera acquisition as extremely positive. Quote: "The grin on my face couldn't be bigger."
David Intersimone described the Idera acquisition as extremely positive. Quote: "The grin on my face couldn't be bigger."
He got a raise? :P
ReplyDeleteAsbjørn Heid that's actually quite probable. During mergers and acquisitions key people often get a rise to keep them from quitting before the new management had time to asses whether they want to keep their part of the acquired company or not and whether they need these people on the future.
ReplyDelete(Been there multiple times myself.)
Russell Weetch, apparently I was wrong - David Intersimone did talk about it :D
ReplyDeleteUsually it is not just raises but a stake in the future company with options or other internal type of dealings. That only happens with top management not the developers pool, they are lucky to have their job. Keep in mind that Idera purchased the company for what is discussed to be over $400 million dollars and the Embarcadero board members are getting a huge profit so that means Idera is not going to be that stupid with the products per say or rush into major decisions in the first 6 months.
ReplyDeleteThe questions I have is what is going to be:
1. Future product pricing along with subscriptions.
2. What products will be discontinued.
3. What is going to happen with the Embarcadero name.
Most of this should be announced a few weeks or sooner after the final sale is made.
I wonder if the folks on subscriptions will get a new welcome from Idera and do I get a free t-shirt ?
David might well be pleased because he can see that Idera are a really good owner for Embarcadero, and that devtools can flourish under their ownership.
ReplyDeleteDavid Heffernan that was the sentiment I got from listening to him live. They want to work their way back into schools. If Rad Studio/Delphi can get into the schools that's a very good thing. :-D
ReplyDelete/future?
ReplyDeleteDavid Heffernan I'm sorry, but that kind of optimism is not allowed in the Delphi community. ;-)
ReplyDeleteDoes anyone know if Idera a public company?
ReplyDeleteNick Hodges Time will tell. If things get better, great. If things get worse, too bad. Evidence over faith in my book.
ReplyDelete+Michael Riley, I don't think that is possible any more. :(
ReplyDeleteUgochukwu Mmaduekwe David I actually mentioned their strategy will include a push to get Delphi into the school system. I believe it can happen.
ReplyDeleteNick Hodges Idera is backed up by TA Associates http://www.ta.com/
ReplyDeleteFrom where I stand things can only get better.
ReplyDeleteLike I said few years back, Embarcadero is pushing hard trying to squeeze money from its customers because Thoma Bravo is expecting to sell them soon enough. It actually took them longer than I expected.
We just have to get used to this kind of shuffling, because this will happen again and again...
I'm happy if David is happy. Bearded sympathy, for sure. ;)
ReplyDeleteJavier Hernández Exit strategy is clearly FPC for me. We made our framework FPC compatible, and may be able to switch if necessary. Top managers don't like to be stuck by one company. The fact that mORMot works with FPC, which is also Open Source and highly maintained, could be taken in account. Perhaps Delphi alone would not have given enough perspective.
Javier Hernández You don't need to worry about exit strategy until you need exit strategy.
ReplyDeleteJavier Hernández Delphi roadmaps are highly unreliable, and extremely vague. I don't expect sudden change in either of those aspects.
ReplyDeleteI've found a recent David I picture, in wikipedia, happy king of Delphi kingdom, proud Idera knight. See https://upload.wikimedia.org/wikipedia/commons/3/30/DavidIofScotland.jpg
ReplyDelete+Michael Riley, what more can I say, let's watch the drama unfold.
ReplyDelete/drama
ReplyDeleteMichael Riley that's what I have been saying for years - give schools licenses - Pascal was always an excellent teaching language as well
ReplyDeleteDidn't Niklaus Wirth originally design Pascal with teaching in mind?
ReplyDeleteMy being a teacher had a decisive influence on making language and systems as simple as possible so that in my teaching, I could concentrate on the essential issues of programming rather than on details of language and notation.
—Niklaus Wirth
Milan Vydareny Language is not the problem, price is. Schools tend to use freely available tools. So in case of Pascal they will use FPC over Delphi or anything else that costs money.
ReplyDeleteDalija Prasnikar I don't disagree with your "price is the problem" analysis. I was simply reinforcing the idea of Russell Weetch that Pascal is an excellent teaching language [because it was designed with that in mind.] I would actually consider trying to teach young people if costs were more acceptable.
ReplyDeleteAbout Pascal in schools, see http://www.pp4s.co.uk/ and more recently https://plus.google.com/114108549026060823053/posts/JqJdCJ2eRgE
ReplyDeleteDalija Prasnikar In the US, it has little to do with price. Despite Pascal being a very good tool for teaching programming, many of the schools are staffed with people who can't see any language that doesn't start with C.
ReplyDeleteJust a few miles from Scotts Valley is a community college which I believe dropped Pascal before Delphi 1. When I took some classes there, they offered Java and C++. It's a school which turns out Cisco techs as a hot commodity. They would be a good place to keep in Object Pascal, but I won't hold my breath.
Most of the staff in such schools are believers that all software should be free.
Bill Meyer Similar thing is happening here. 25 years ago it was Pascal everywhere, now it is C/C++, Java, Python. But even when you have Pascal classes Delphi is nowhere to be seen because of price.
ReplyDeleteDalija Prasnikar But now, even FPC won't bring it back in this country.
ReplyDeleteI've said it before and I'll say it again: Delphi doesn't have to be wildly popular, it just has to be profitable, which apparently it is, given that it is a major part of EMBT, and what Idera just paid.
ReplyDeleteMichael Riley The Department of Basic Education in South Africa started teaching Delphi exlusively in all our schools in 2014. Unfortunately this decision was not well received at all by the tech industry locally, referring to Delphi as a "stone age language" amongst other (less flattering) things. http://www.webaddict.co.za/2013/10/09/south-africa-education-department-bans-open-source-software/
ReplyDeleteMichael Riley If David says this he has no idea what he's talking about unfortunately. Conferences were held in the late '90s to discuss what would be the future language(s) of choice for educational use. As one paper I read put it, "The only consensus that was reached was that it wouldn't be Pascal".
ReplyDeletePascal at the time was the language used for Advanced Placement Computer Science. It was replaced way back in '98 or '99 with C++ for a brief period, and then switched to Java.
You can't possibly, 15 years later, talk people into going back to what they left 15 years earlier. On top of that, open source has moved into education in a big way. A closed-source language in 2015 is going to go over like a lead weight. As computer guru and education advocate John "Maddog" Hall likes to put it during conferences, open source languages teach you twice: once to learn the language, and then later students can look "under the hood" and learn how the interpreter/compiler/VM work. They can also learn about contributing code and working with others in a collaborative fashion.
David saying they want to get back into school is like how, during the CodeGear days, they were talking about adding special features for parallel programming right into the language. Eight years later and it never happened. They might as well want to get into Google and Facebook too. This is a wish, not a plan. It gets everyone excited, but they have no means of actually delivering on it.
Remember what happened when South Africa tried to make Delphi the mandatory language for their school system? Teachers rose up in protest against it and professors wrote scathing blogs about the idea. There was so much unrest that SA dropped the idea.
I don't know about anyone else, but I talked to the professor leading the charge and I got ahold of the policy paper the SA Dept of Education put out to back their plan. It was a nightmare. It claimed that Delphi was very popular, "especially in the United States and Western Europe". Of course, those are the areas it's least popular. It also seemed to pick up lots of dubious facts straight from the Delphi Non-technical forum. It stated that because Java is open source, "there might not be any support for it". Java. The language guided by one of the biggest tech companies in the world, Oracle. Java, one of, if not the, most used languages in the world. That Java.
Did anyone at EMBT help them put together this laughable document? If yes - they're clueless. If no - why didn't they try to help sell the idea? Did they review this document and offer any feedback?
So, as much as Nick might think this is pessimism, I think it's simple extrapolation - the one time EMBT tried to get back into education, the entire process was mismanaged and allowed to become a PR nightmare with teachers openly revolting against it (including posting "What year is it?" memes) and SA putting out a ridiculous white paper that suggested whoever wrote it didn't even know what Java was. Delphi came out of it with a damaged reputation.
Damon Runyon once wrote, "The race is not always to the swift, nor the battle to the strong, but that's the way to bet."
There's no sign that EMBT can move mountains and get Delphi into education when they did so poorly the one time they tried. If Idera thinks they can get back into education, then the new owners are still stuck in the '90s - actually the 80s now. Kids today pop in a Ubuntu image on a flash drive and within minutes they've downloaded half a dozen compilers and interpreters. Then they watch free online courses and YouTube videos, read some open source books and start hacking. They don't spend money on proprietary languages and they don't use software that only runs on Windows. Their grandpa runs Windows. The really cool kids are running Arch Linux today.
Michael Riley "David I actually mentioned their strategy will include a push to get Delphi into the school system." That's not a strategy, that's a wish. That's no different than "I plan to alleviate my financial problems by getting more money". "Getting more money" is a goal, not a plan.
ReplyDeleteDalija Prasnikar "You don't need to worry about exit strategy until you need exit strategy. " By the time you need the exit strategy it may be too late. This scenario is something we should have been talking about as a community the day CodeGear was purchased; we knew it was going to be "flipped" sooner or later.
ReplyDeleteIn 2008 I adopted the simple rule, "When all else is equal, choose cross-platform. If all else is still equal, choose open source". When 2010 came and my Win XP imploded and I had to decide to reinstall, upgrade to Win7 or switch OSes, I was able to evaluate all of my options and ended up making an almost painless transition to Linux because most of the programs I was using by then had Linux versions. Today I know that I can leave Linux as well if I wish. Two years of preparation had removed me from vendor lock-in.
Arnaud making his products FPC compatible was a good idea, both for himself and others.
Nick Hodges Actually you were always arguing with me about it being popular. ;-) But being profitable is a terrible metric for a language. Being good at what it's designed to do, making its users more productive, is the metric I want a tool to aspire to.
ReplyDeleteReminds me of when I was working in logistics. One company I briefly worked at used to ask their salesman, "What's the lowest savings the customer would be willing to accept and still use us?" They'd negotiate in bulk with carriers but pocket as much of the profit as they could, and they'd go with carriers who'd give them kickbacks. In that case, lowering your costs would make them less money.
A co-worker left and started his own company, which I joined as employee #1. We made money based on how much we saved you, period. No kickbacks. We couldn't make money unless we saved it for you. We priced our software tools as well so that they paid for themselves in six months. This way our goals and the customers' goals were aligned. You knew that every day one of us was thinking about how to save you money.
If EMBT sold for at least $450 million then Delphi was not a major part of the company, even if it doubled in value since the CodeGear purchase.
Finally, how much money was made from existing customers vs. new ones is a most important metric.
Bill Meyer "Despite Pascal being a very good tool for teaching programming, many of the schools are staffed with people who can't see any language that doesn't start with C."
ReplyDeleteIt was a good tool, back in the 1980s. But it's 2015 today. Kids have Macs and Android tablets today and the cool hacker types all run Linux. Delphi still runs only on Windows. It's also gotten far more complicated than it was in the Turbo Pascal days (and obviously more expensive). It's also not used much nowadays in business, so students aren't going to be excited about learning something that they don't feel would be useful to them. If you learn Java or C++ in high school and go on to college to take computer science, you know you'll see those languages again. You're not going to see Pascal again, so why not learn the same language you'll be using in college? You also can't find Pascal textbooks or learning materials. Kids can't go into the bookstore and pick up a Delphi book or magazine. The AP computer science exam is administered in Java; the last year it was administered in Pascal was 1999.
Basically, the C-based languages won, and they won a long time ago. It's simply a matter of practicality at that point.
What argument would you make such a school, given all the drawbacks of using a much less popular, and more expensive, language? Heck, South Africa had to commission its own Pascal textbook! What benefit offsets all that hassle? Again, not a conspiracy, just practicality.
David's job is to evangelize. How many times have you heard him make a negitive comment? The new company needs him to say positive things.
ReplyDeleteAlso, be is pretty high up (was or is at least a VP?)... Stock? Raise? Bonus Bucks? Retirement plans?
I'm sure his statement was fact.
But let's see if we are smiling in a year or few down the line. We invest time and code, and bank our businessnes and liveleyhoods on a product that keeps getting sold down the river. Its a great product but also a hot potato that few seem to want to get caught holding onto.
Joseph Mitzen Having seen the grotesque process of using C++ in an entry level class, I am confident that the students would have gained more real understanding using Pascal. Yes, C won the war, but it still has the nastiest syntax irregularities of any supposedly modern language. And C++ containing C, retains support for all that. And don't even get me started on recursive text preprocessors.
ReplyDeleteIt's actually quite possible, and preferable, to learn C++ without any mention of C. The standard library has a broad selection of string and container types, algorithms, and so on. No need to use pointers for a long time. No need for memory allocation for a long time. It's quite possible to teach C++ in a good way.
ReplyDeleteI think though that if I were wanting to teach programming to people moving beyond environments such as Scratch, then I'd opt for Python as a start.
Joseph Mitzen If you are using Delphi today, and finding yourself worrying about exit strategy then it is already too late.
ReplyDeleteI wonder why people spend 40+ minutes (just look at the posting times) writing in a single discussing thread, if Delphi is so unimportant to them. At least Delphi is important as a strong force to move you away from the real important things in your life, no?
ReplyDeleteI've seen this over and over since the psyco Delphi Hater guy: The energy and time they waste just trying to diminish Delphi actually says quite the opposite.
Don't take me wrong. I mean.... Man, I hate plain C and I really think it should be eradicated from the face of the earth like a plague. I've been working so much that sometimes I wish I could have at least a few more hours to spend with my family, away of computers. Now, imagine that I will join some C discussion group and waste 40 minutes of my precious life saying that C is crap, C is unsafe, C is old, C is whatever. Man, the day any of you see me doing this, you are granted the right to put me in a mental institution.
ReplyDeleteAlexandre Machado There are entire websites, such as LOLPHP, dedicated to making fun of PHP. People criticize Delphi because it is important to them.
ReplyDeleteLiving in an echo chamber where everyone reinforces each other's viewpoint is far more dangerous for... well, anything, including computer languages.
Let me give you an example: This is a real quote from a commenter on The Register's article about Idera buying Embarcadero:
"Delphi is much more widely used than many people think - I have heard reasonable estimates, which I believe, that it's about on par with Python."
This is what happens when you have a community with no outside input. Statements like this are far from rare, unfortunately. Heck, a former product manager:
" ...I think there are a lot more Delphi developers out there than people know. Is it more popular than C/C++? Maybe not, but then again, maybe it is. No way to know. "
And even a current product manager:
" I'm convinced that in the business world Python has a fraction of the Delphi influence"
(probably the inspiration for the original commenter)
If we as a community don't even know where the language stands in the marketplace, how can we make reliable decisions about anything else related to Delphi?
We could do with a lot more gadflies, not less.
“An acquaintance merely enjoys your company, a fair-weather companion flatters when all is well, a true friend has your best interests at heart and the pluck to tell you what you need to hear.”
― E.A. Bucchianeri, Brushstrokes of a Gadfly
Javier Hernández Java is not faster in most cases. Asbjørn Heid showed that Java x64 was faster in that particular algorithm, probably because of a better floating point code generation. There are many other aspects there:
ReplyDelete- Not all real world applications are affected by floating point calculation speed. Even when they are, other factors are much more important. I've already worked with engineering and banking applications and floating point performance was not even close to be the bottle neck.
- In many other areas Delphi still kicks Java (and even more C#) ass in terms of performance. One of the banking systems that I developed as part of a team, have beaten the BEST Java commercial application in the country by a factor of 2! Our multi-threaded Delphi server could handle 2 times more XML messages per hour than the faster Java software (load information from the database, create a formatted XML message, send it to a remote server using a TCP connection, receive the response, validate and parse the response, and finally save the information back to the DB).
I was just laughing at the Java "Oracle certified" gurus faces when they discovered that they were beaten by an "your granpa uses" Delphi application. LOL!
Don't fool yourself with those limited tests. It is really HARD to beat a well written Delphi application.
Joseph Mitzen Really? I don't want to waste any more of your time... Those LOLPHP guys probably don't have a life. I do. If you think that Delphi is crap. Just stop using it!
ReplyDeleteJavier Hernández Yes, I'm serious. I can't reveal the names of the involved companies, but I can give you the email of the TI manager of the bank. He can confirm exactly what I told you here. Besides faster floating point in x64, what else you get?
ReplyDeleteHum... and if I'm not mistaken, A. Bouchez mormot server also kicked some Java and C# servers out there, didn't it?
ReplyDelete/unsub
ReplyDeleteJavier Hernández so what you are saying is that replacing an internal library (What RTL is actually) with another, which has better performance, somehow is "not allowed" in your book? Use custom assembly code in a compiler that gives you that option is also something "not allowed". Hum... I see.
ReplyDeleteAlexandre Machado
ReplyDelete>Java is not faster in most cases.
I haven't seen a benchmark that's shown this, so I don't know how you can say "most cases".
Another real story: on another site someone chastised me for pointing out that Java was faster and insisted that it was impossible. I posted the saga of the SciMark benchmark, in which both Java and C# beat Delphi.
Anyway, this poster must have scoured the evidence for proof I was wrong and then came back with a scientific paper with benchmarks that proved Delphi was faster! How foolish I was!
I was not familiar with the journal in question so I went and checked it with one of several blogs that keep track of fake journals that will publish any paper for a price or were only set up to publish the papers of one person. Sure enough, this journal was listed there.
I went and read the paper next. One of the benchmarks was printing hello world! Another was executing an empty loop! I searched for other papers from this author and sure enough every paper they'd had published was in this one journal, and the best one looked like a project I did in high school.
Apparently since it told people what they wanted to hear, they didn't feel they needed to read it before citing it.
> +Asbjørn Heid showed that Java x64 was
>faster in that particular algorithm, probably
>because of a better floating point code
>generation.
Modern Java is faster because its JIT compiler is optimized far more than Delphi's desktop compiler, which dates back to DOS days and consists of over a million lines of undocumented C code (source: Barry Kelly on Stack Overflow). JIT compilers can also perform classes of optimization that standard compilers can't. Finally, and most importantly, Delphi's compiler is based on a single pass compiler design. This is why it is so fast at compiling, but it also eliminates entire classes of optimizations.
>- In many other areas Delphi still kicks Java
>(and even more C#) ass in terms of
>performance.
Post the benchmark.
But even if it did, I should sure as heck hope it should since it doesn't target a VM. When comparing apples to apples, I haven't seen a benchmark in which it's achieved better than 50% of C++'s performance.
> Our multi-threaded Delphi server could
>handle 2 times more XML messages per
> hour
Was the Java software multithreaded?
>Don't fool yourself with those limited tests. It
>is really HARD to beat a well written Delphi
>application.
Delphi users claim this but Delphi users don't post the benchmarks so it never has any effect. If you want to prove something nowadays, you have to post the code (and then put it on Reddit or Slashdot). Are you aware of a benchmark that can demonstrate this? If you are, you have a PR coup that EMBT has missed.
Yet even then, one would expect a bare-metal language to be faster than a VM-based language. It's more impressive when Java beats Delphi than if Delphi beats Java unless the result is very significant.
Now if you can convince a C++ developer (what one uses in the real world when speed is the only priority, which it seldom is) that it's hard to beat a Delphi application, I'll eat my hat. The selling points of Java and C# are primarily speed of development, reliability, rich standard libraries, vast ecosystems and plenty of trained developers. Those are the points Delphi would need to compete against with an entity considering Java or C#.
Alexandre Machado
ReplyDelete>If you think that Delphi is crap. Just stop
>using it!
It never ends in this place... if you think Delphi isn't crap, why don't you correct other people's misconceptions? Why can't we have a decent conversation without people feeling they're being personally assaulted? It's just a programming language. It doesn't have feelings.
There's a popular website that flatly states that all programming languages are crap, some are just less crappy at certain tasks than others. :-)
Alexandre Machado
ReplyDelete> so what you are saying is that replacing an
>internal library (What RTL is actually) with
>another, which has better performance,
>somehow is "not allowed" in your book?
I'm not familiar with internals of Mormot but It sounds like he's saying that you're paying thousands of dollars and they still have to tweak the shipping RTL to get good performance. If that's the case, then the performance of Mormot is a credit to Arnaud, not to Embarcadero.
>Use custom assembly code in a compiler
>that gives you that option is also something
>"not allowed".
Now this I can address. Firstly, that would be considered a bad idea nowadays since the code wouldn't be portable across architectures. Secondly, hand-written assembler in 2015 is considered unnecessary because compilers are able to produce machine code better than all but masters of assembly language. Third, if you have to add manual assembly language into a compiled language it's a huge red flag that your compiler is terrible. What are you paying money for if you have to hand-write the assembler yourself?
Lastly, people today step down to C++ (and sometimes C) when they have bottlenecks, not assembler. Obviously C++ is a lot easier to write and debug.
You might be getting speed, but if you have to alter the RTL yourself and then hand-write assembly, your productivity is going to be through the floor. Other tools would have been done much sooner, allowing more time to optimize bottlenecks. Heck, if would be quicker to write the whole thing in C++ because you wouldn't need the tweaks, and it would be cheaper too.
So no, if indeed the RTL needs patching and assembly needs to be written, that's not a compliment. That's an incredible drain of developer time, which is more costly than computer time. It reminds me of Jeff Atwood's blog post, Hardware Is Cheap, Programmers Are Expensive:
http://blog.codinghorror.com/hardware-is-cheap-programmers-are-expensive/
"Hardware Is Cheap, Programmers Are Expensive" and one paragraph above I also read "if would be quicker to write the whole thing in C++ because you wouldn't need the tweaks, and it would be cheaper too"
ReplyDeleteSo, why aren't you writing your own programs in C++? Do you have a SINGLE program in Delphi? If yes, why? If not, what the hell are you doing in a Delphi group? You don't have friends?
+Alexandre Nothing about this group says that members must not be critical. Please stop the abuse.
ReplyDeleteDavid Heffernan Forgive me, but which abuse? I'm genuinely trying to understand the reasons why some developers spend so much time and energy criticizing something that, at least in theory, they don't use (or shouldn't be using according to their own point of view). Am I allowed to ask this?
ReplyDeleteAlexandre Machado Your previous comments have been insulting and abusive. If you disagree with reasoned criticism, make your argument, or let it go.
ReplyDeleteI agree with David Heffernan. Keeping a cool head and making statements without added drama or innuendo helps further debate, instead of derailing it. Stay on topic.
ReplyDelete/sub (again)
ReplyDelete