Embarcadero Delphi is sold!

[x2nie]

I am thinking: If Idera only care about the database product, will they drop Delphi soon?
IMHO, Microsoft usually only focus on one aspect while they buy a company. such as Nokia, they focus on hardware production, dropping powerfull Nokia Symbian OS, and all user base (community) may be ignored then since. I assumed the same things may occurred to Delphi
[/size]"Idera, a provider of monitoring and administration tools for Microsoft's SQL Server, is to acquire Embarcadero from private equity company Thoma Bravo. ..."


http://www.theregister.co.uk/2015/10/08/embarcadero_database_and_app_dev_company_sold_to_idera/

[jwdietrich]

That it very interesting.

It is now more important than ever to educate the huge community of Delphi developers that Lazarus / FPC is a serious alternative, which can help them to save their investments in coding, at least in part.

[rvk]

I am thinking: If Idera only care about the database product, will they drop Delphi soon?

This might shed some light on the matter:
http://community.embarcadero.com/article/news/16201-customer-and-partner-welcome-from-idera

I am Randy Jacops, the CEO of IDERA, Inc....
...
For Embarcadero’s development tools, we acknowledge this is a new opportunity for IDERA. I have seen some chatter on our community forums and received direct questions from customers on IDERA’s intentions for the product line. I wish to emphasize the tremendous value we perceive in development tools. As applications proliferate across multiple channels and platforms, the need for tools to build the next generation of applications will soar. Concepts such as the Internet of Things or proximity awareness in applications did not exist several years ago. Today, everybody recognizes the opportunity in building applications for equipment that simplifies and advances society. We believe the winners in the application development market will focus on the developer and differentiate with speed to market, ease of use, and pace of change. All of these things perfectly align with IDERA’s view of product and customer success.

I'm sure they're not just going to drop Delphi.

[BeniBela]

• For-in loop support
• Generic records, arrays and procedural types

Okay, those might be useful.

• Enumerators in records

Just a consequence of for-in-loops

• Support for class and record helpers
• Support for type helpers added

As long as only one of them can be active that is pretty much useless

• Support for non-local goto's
• True and False are not keywords anymore

That does not sound like a good idea

• Better support for Delphi-compatible classes - this includes newer Delphi versions too.
• Support for the Objective-Pascal dialect
• Scoped enumerations
• Basic ISO Standard Pascal support
• Support for &-escaping of keywords
• Support for univ parameters in MacPas mode
• CExtended floating point type
• Using the function name as alias for loading its current result value in MacPas mode
• nostackframe forbidden for Pascal subroutines

Those are just compatibility changes

If you do not use the things it became compatible to, they are irrelevant

• Overload modifier must be present in the interface
• Default values in implementation but not in interface/forward declaration
• Default values are now properly typechecked

Those have really annoyed me.

Now FPC cannot compile my old Delphi 4 code anymore

• Sealed and Abstract classes
• Advanced record syntax
• Custom deprecated messages
• Constref parameter modifier
• Support for nested procedure variables
• Anonymous inherited calls
• "strict protected" visibility modifier
• "strict protected" and "protected" visibility modifier in extended records
• "static" directive on non-class methods
• "static" directive on class operators
• Classes implementing forward-declared interfaces
• Comparative operators can have any result type
• For-in loop variable cannot be assigned to anymore

Such minor changes barely matter

• SAR intrinsics
• ROL/ROR intrinsics
• Bitscan intrinsics
• Boolean16, Boolean32 and Boolean64 types

Delphi 4 could just have those in a utils unit

• AnsiStrings are now codepage-aware

Too bad Lazarus ruined that  by removing the ansi codepage

• Variant overload preference for string types
• Variant conversion preference for widechar
• Conversion preference of pansichar to various string types

Too bad it is not using utf8string
 

• Casting integer variables to floating point

Great, now I need to check all my casts if they still work in fpc 3

Big advantage of Delphi 4: Working source continues to work for decades and there are no random changes that break everything.

[Thaddy]

Big advantage of Delphi 4: Working source continues to work for decades and there are no random changes that break everything.

And some more flaming.
Your D4 code works perfectly fine in FPC 3 given some minor changes that won't break the D4 code or would have been broken by higher Delphi versions. D4 is extremely buggy.
And D4 was hardly a shining example of Quality. It has an even number  . 4,6,8 were disasters. 3,5,7 were highly stable and D7 is still highly regarded and much used.
btw: I still have D4 installed and it get used now and then, as I have 2,3,5,6,7,2006,2007, 2010, XE and XE2. ( for KOL I wrote the system unit replacements for D4). I also have D1, but not installed anywhere.

[BeniBela]

Your D4 code works perfectly fine in FPC 3 given some minor changes that won't break the D4 code

But I have to check/test every line

D4 is extremely buggy.

I has always worked fine for me

btw: I still have D4 installed and it get used now and then, as I have 2,3,5,6,7,2006,2007, 2010, XE and XE2. ( for KOL I wrote the system unit replacements for D4). I also have D1, but not installed anywhere.

I only have D1, D3 and D4

And XE or something, but not installed anywhere


D4 is the best of them

[garlar27]

Boring off-topic discussion!!!

Although I'm not going to use Delphi again (except if I'm paid for), IOW they don´t drop Delphi and focus their effort to make it better instead of just make money, because money comes as a natural result of an activity focussed in achieve a "natural" goal (in case of Delphi would be a better product the same goal of FPC and Lazarus) instead of focusing on money earning. If you move your focus from product or service to profit you will start to loose clients because your product/service is geting far behind your competitors.

This is something that some finance companies will never understand.

[lagprogramming]

Boring off-topic discussion!!!

Although I'm not going to use Delphi again (except if I'm paid for), IOW they don´t drop Delphi and focus their effort to make it better instead of just make money, because money comes as a natural result of an activity focussed in achieve a "natural" goal (in case of Delphi would be a better product the same goal of FPC and Lazarus) instead of focusing on money earning. If you move your focus from product or service to profit you will start to loose clients because your product/service is getting far behind your competitors.

This is something that some finance companies will never understand.

    As an analogy, you share the same point of view with Toyota in the 80's. They considered that a good car will sell by itself, without much investment in commercials(publicity). They were right. In the meantime, the world has changed. Nowadays, there is an argue over diesel emmisions. Although the cars were not as good as competitors, a manufacturer managed to sell lots of them, for years.
   In the example above, all manufacturers focused on sales, but with different approaches. My point is that even if you have a great idea, even if you put that idea in practice, it's not everything that matters. There are additional things that put together, sometimes may have a bigger impact.

[Cyrax]

• Support for class and record helpers
• Support for type helpers added

As long as only one of them can be active that is pretty much useless

Huh? There can be many class/record/type helpers active any time.

• AnsiStrings are now codepage-aware

Too bad Lazarus ruined that  by removing the ansi codepage

Eh? It is there, if you want to use it. Lazarus uses UTF8 as default codepage.

• Casting integer variables to floating point

Great, now I need to check all my casts if they still work in fpc 3

I think that applies only explicit casts.

Big advantage of Delphi 4: Working source continues to work for decades and there are no random changes that break everything.

Big advantage of FPC/Lazarus : it isn't closed source and YOU can fix bugs in it!  . Or report them and get them fixed in due time, unlike with Delphi...

[JuhaManninen]

Too bad Lazarus ruined that  by removing the ansi codepage

Eh? It is there, if you want to use it. Lazarus uses UTF8 as default codepage.

Unfortunately the system codepage is gone when the new UTF-8 system is used. It swithes String to use UTF-8 but String = AnsiString in FPC, thus AnsiString will be UTF-8, too.
I understand there is code that relies on data having system encoding. We must support also such code.
I will collect energy to update the wiki page and then open another thread for discussion. I will need help later to update the wiki more.

[garlar27]

    As an analogy, you share the same point of view with Toyota in the 80's. They considered that a good car will sell by itself, without much investment in commercials(publicity). They were right. In the meantime, the world has changed. Nowadays, there is an argue over diesel emmisions. Although the cars were not as good as competitors, a manufacturer managed to sell lots of them, for years.
   In the example above, all manufacturers focused on sales, but with different approaches. My point is that even if you have a great idea, even if you put that idea in practice, it's not everything that matters. There are additional things that put together, sometimes may have a bigger impact.

"To marketing or not to marketing. That's the question?"
There are plenty of notorious storys of success with excesive marketing and with a complete lack of it, and so it goes for big fails.

IT people do the IT stuff, Sales people do the sales. And people with knowledge in both areas can be the "mediators" between them.
Pure black and pure white exists only in the color palettes of a paint program. The real life is full of colors, tones, shades, shapes and textures.
I know you already know that, but sometimes when someone gets passionate in a discussion those things become blurrry.

[ykot]

Some more discussion on this topic is on Embarcadero forums. Reading through that discussion, some bad thoughts come to mind: so Thoma Bravo bought Embarcadero, then bought CodeGear for price X, then did something to make the whole firm cost much more, then sold it to Idera for price K * X being much higher. The disturbing thought is: why would you want to sell a firm, which is a lucrative enterprise with big potential? Perhaps Thoma Bravo saw the beginning of Embarcadero's demise and sold it before it's too late?

Some more troublesome logic: owner A (Thoma Bravo) with big budget owns firm X (Embarcadero). Now, that owner A decides to sell firm X to another owner B (Idera), which doesn't have enough bugdet to pay for it, so it takes a big credit. In your opinion, what is better for firm X, to have an owner A, which has big budget, or to have an owner B, which besides not having sufficient budget now also is in a big debt? In such case, which of the owners A or B would more likely have budget to invest in firm X to continue its growth?

In other words, for me it looks like Embarcadero will have some difficult time as new owner Idera, apparently, seems to have limited budget when compared to previous owner, and also needs to pay for its debt.

Can someone please come up with a more positive/optimistic logic about this and explain it?

[BeniBela]

Huh? There can be many class/record/type helpers active any time.

According to what I have heard, only one for every type is allowed

I think that applies only explicit casts.

All my explicit casts then.

There should not be any implicit float to int casts

Big advantage of FPC/Lazarus : it isn't closed source and YOU can fix bugs in it!  . Or report them and get them fixed in due time, unlike with Delphi...

I tried that. Turns out it is more efficient to just rewrite everything from scratch.

I will collect energy to update the wiki page and then open another thread for discussion. I will need help later to update the wiki more.

Better implement something where we get string = utf8string <> ansistring

Can someone please come up with a more positive/optimistic logic about this and explain it?

If Delphi fails, they all come to FPC?

And FPC can stop to implement compatibility-nonsense ?

[taazz]

Some more discussion on this topic is on Embarcadero forums. Reading through that discussion, some bad thoughts come to mind: so Thoma Bravo bought Embarcadero, then bought CodeGear for price X, then did something to make the whole firm cost much more, then sold it to Idera for price K * X being much higher. The disturbing thought is: why would you want to sell a firm, which is a lucrative enterprise with big potential? Perhaps Thoma Bravo saw the beginning of Embarcadero's demise and sold it before it's too late?

You are not looking at the forest. Thoma Bravo is a investor not a technical oriented person. What is the life cycle of a successful investment? Buy low, raise its value sell at the highest possible point to maximize revenue. Any thing else outside that logic is conjecture and probably a waste of time.
there are a couple of questions to be asked.
1) Is the current value the highest that the current company can go?
   Probably not there is at least 20% more revenue to be made otherwise it would be stupid to buy do you think that the new owners made the decision to buy lightly with out any plan to make a profit?
2) Can the experts used to extend the market?
3) How many customers will be lost if the company goes to support mode only? 

after answering those questions you will have you answer. My assumption is that the new owner is looking to expand its market. When you reach that point there only 2 directions you can go forward (aka closer to the consumer) or backwards (aka closer to the producer). Apple choose forward (opening apple stores in various countries) and Idera chose to go backwards. The only thing certain is that a step of that magnitude is usually well thought out. The thing with the backward move is that it can't be done easily its not something any one will choose to do unless he intents to control the specific market. That drives me to the conclusion that idena will not simple give up but from where I'm stunting delphi is not part of Ideras current goal its something extra a perk something that can be easily discarded if required.

Some more troublesome logic: owner A (Thoma Bravo) with big budget owns firm X (Embarcadero). Now, that owner A decides to sell firm X to another owner B (Idera), which doesn't have enough bugdet to pay for it, so it takes a big credit. In your opinion, what is better for firm X, to have an owner A, which has big budget, or to have an owner B, which besides not having sufficient budget now also is in a big debt? In such case, which of the owners A or B would more likely have budget to invest in firm X to continue its growth?

In other words, for me it looks like Embarcadero will have some difficult time as new owner Idera, apparently, seems to have limited budget when compared to previous owner, and also needs to pay for its debt.

Can someone please come up with a more positive/optimistic logic about this and explain it?

liquidation  is the only goal of any investor.

[taazz]

If Delphi fails, they all come to FPC?

And FPC can stop to implement compatibility-nonsense ?

More like they all go to .net and java. Only the most stubborn (like me) will come to fpc.