Lars Fosdal: VS/C#6 is out-delphi-ing Delphi. I hope EMBT is paying attention. Currently…
Posted by jpluimers on 2015/05/01
Interesting, Lars Fosdal writes:
VS/C#6 is out-delphi-ing Delphi.
I hope EMBT is paying attention.
Currently live at http://channel9.msdn.com/
I already liked the 7 minute C#6 video from last year. This is even better.
–jeroen
via: VS/C#6 is out-delphi-ing Delphi. I hope EMBT is paying attention. Currently….
The Wiert Corner - irregular stream of stuff 
Kmorwath said
You still make the big mistake to consider C# the competitor of Delphi. It was the same mistake made aiming it at VB that made Delphi quite irrelevant in the development landscape. Delphi must target C++ with the aim of being almost powerful as C++, but far easier to learn and write, but able to almost cover the same “development surface”. Keep on aiming it at C#, Python, whatever, and it can be only a loser. Face it, those choosing C# & C. want a language able to protect low-skill developers from their own mistake and quickly produce so-so applications. There are really no great applications written in C# alone. Keep on firing in your feet, Delphi “developers”, you’re not going far…
A. Bouchez said
+1 for my own use of Delphi: a replacement of C and C++, for critical server applications, and efficient RAD clients.
But do not forget that a vast majority of Delphier are in fact only using the RAD approach of it, far away from a true OOP design.
From the official Embarcadero marketing point of view, it sounds like if they are putting Delphi as a RAD tool (“write without any code”)…
So perhaps your comment should be addressed more likely to the current Delphi staff…
Marjan said
Who said we were comparing languages?
I certainly wasn’t.
I am drooling because the C# language and Visual Studio teams show that they are doing their utmost to help developers be awesome by tweaking the language and the IDE to make both a help rather than a hindrance and “actually make things works” (to quote Mads Torgersen.)
In its infancy Delhi revolutionized windows application development by making things much, much, much easier for developers.
Currently, both the language features and the IDE are sadly lagging behind and lacking in how they help developers do their job of producing applications that make money for them and/or their employers.
I will never understand that EMBT do not address something as basic as the fact that you can’t even compile an entire project group without out-of-memory errors unless you install a third party add-on. This started in XE2 IIRC and has gotten worse in XE6. And that’s just one of the many daily frustrations I have with the IDE and the compiler(s).
Horses for courses and all that, sure, but I just wish EMBT would for once produce a horse that could actually run a course…
jpluimers said
I totally agree.
KMorwath said
No. It was Visual Basic (1991) that revolutionized Windows development making things much more easier. Delphi came only later 1995, and wit a 16 bit version only! (because in 1991 Borland was obsessed with purchasing dBase…) and copied (it copied well, but MS was already ahead). It wasn’t Delphi to introduce the RAD approach. It was VB. Delphi added a true OOP language and a true compiler. Which made it able to compete with C++ to deliver powerful Windows applications VB could not deliver, yet VB was much easier to learn and use, and that’s why it remained more popular than Delphi even at the latter peak.
Aiming Delphi at VB developer, Borland aimed it at the wrong target. Most of them weren’t interested in OOP, “full” API access, native compiler, etc. They were only interested in “quick and dirty” applications, and, well, even too many Delphi developers used Delphi exactly the same way (probably because ActiveX control required registration… Delphi controls didn’t…).
Given the price difference, if you aim the product at developers who aren’t interested in the high-end features (and when those features became less and less, and buggier and buggier), they will buy the other product. Cheaper, backed by a far larger company, with much more stuff already in the box, or the framework…
The early success of Delphi made Borland complacent… and when it found that success was crumbling, it acted in several schizofrenic ways – unable to assess reality, and trying in sequence several ill-fated “alternative realities” (Linux, .NET, OSX, mobile, and counting…)
The IDE is a disaster, sure. The Galileo IDE is ill-born, and the culprit is…. .NET. They needed to support C#, .NET and with scarce resources took the easy way, and the result is the actual IDE. The compiler is also a disaster, because of course aiming it at VB/C# developers made the compiler quality “irrelevant”.
Language features? Beware of a “me too” approach for lack of a clear direction – a language must evolve within a clear idea of what it is, and where it wants to go. Which “development envelope” it should cover. A “Jack of all trades, master of none” is of little use but for the laziest developers unwillingly to learn new languages for different tasks and hoping to use just one for everything.
C# evolution is backed by controlling the .NET IL, virtual machine, run time and framework. And after twelve year, it’s clear it’s not designed to take the place of C++ for heavy applications. MS has put much more stress on C++ for developing many types of applications, including the mobile ones.
Does Delphi want still to be a fully compiled language? Should it still be able to fully support Windows (any Windows…) without VMs or the like? Should it still be able to empower the developer with full memory control and access? Delphi probably needs to do less and better, not as much as it can and so-so, if not really bad. Why? Because behind C# there is the huge and mighty MS, and behind Delphi the little, lame Embarcadero.
Otherwise, do what RemObjects did. Put some Pascal varnish over the .NET IL, VM and framework, add as many new keywords as you like, put everything into VS IDE, bake at 180° for one hour, and you’re done.
Marjan said
So Delphi can’t even claim the revolution? Well whaddayouknow…
I agree that EMBT (and its predecessors) are “flailing” and playing a “me too” catch up game that they are never going to win because they just don’t have the resources to do it all well enough to be of enough value.
For me using Delphi no longer is a choice, but a consequence of my job…
Wouter van Nifterick said
While many Delphi developers want the with-statement to be removed from the language, I see the evil twin brother being introduced into C# at around 19 minutes into the video. The “using static XXX” language feature looks worse than the with-statement.
Imagine the horror of unit-wide with-statements in Delphi! :)
sglienke said
Totally wrong. with introduces a closer scope than any local var, parameter or member. The new static using cannot do that. It will not interfer with any existing scope/code. If you have a WriteLine method inside your class it will still call that method if you write using Syste.Console. A with statement in Delphi interfers with the scope – and that is one of its dangers.
Marjan said
I am drooling…
Delphi / EMBT eat your heart out.
abouchez said
Some of those are similar to the “Smart Pascal” dialect enhancements, especially about expressions and properties.
See http://en.wikipedia.org/wiki/Smart_Pascal#Linguistic_differences