Array resize bug in beta build 530 - page 2

 
cyclops993:

Not only a public notification alert; also a public alert. What else is the point of a forum?

If there's no public discussion and notification then the service desk waste time getting the same bug report multiple times, and users waste time diligently investigating and double-checking something which has already been reported.

(See also https://forum.mql4.com/56885/page20#861740)

 
I don't think metaQuotes provides backward compatibility for ex4 files in this sense. Should the .mqh* codes work after recompiling with the new compiler then I'm fine with that. But thats just me... I don't know what the industry standards are, however, (imo) at some point legacy cannot be supported by any company.
 
ubzen:
I don't think metaQuotes provides backward compatibility for ex4 files in this sense. Should the .mqh* codes work after recompiling with the new compiler then I'm fine with that. But thats just me... I don't know what the industry standards are, however, (imo) at some point legacy cannot be supported by any company.

MeraQuotes tried abandoning legacy code. That's what MT5 is, and what a success that turned out to be.

Abandoning existing compiled code would be commercial suicide, and MetaQuotes clearly aren't trying that. The new version is clearly trying to support legacy code, but there are some fundamental holes at the moment.

 
ubzen:

Are you creating these Smiley in photo-shop. !@#$% :). Anyways, (today) I'm the devils advocate, tomorrow I might be the one B!@#$%.

Ps: If it makes anyone feel better..... English_Rules !!!

No, not creating them, they are the Skype ones I "borrowed" . . . they are just animated GIfs, you can also copy and paste them just as you would text or insert them as you would any other image.
 
cyclops993: MeraQuotes tried abandoning legacy code. That's what MT5 is, and what a success that turned out to be.

Abandoning existing compiled code would be commercial suicide, and MetaQuotes clearly aren't trying that. The new version is clearly trying to support legacy code, but there are some fundamental holes at the moment.

Well one things* for sure. Someone wouldn't be able to compile old codes with the new compiler (for example variable.name) with a dot_separator wouldn't work in the new compiler. metaQuotes being a pretty good developer would probable get the ex4 support correct (if that's indeed their intend). But I don't see Microsoft (for example) supporting programs for WindowsXp. I have quite a few programs which wouldn't install on windows7 even in Xp/Compatibility_Mode.

metaTrader5 and metaTrader4 offer different features and thats why I believe 4 is still around. mt5 cannot be accepted by the exchanges/sectors which MQ wanted to enter with the features of mt4 included. I don't believe that mt4 is around because of commercial suicide if abandoned. Should mt4 become un_supported, most would just move_onto mt5. <--#1 reason most would is because of the online_community.

 
ubzen:

Well one things for sure. Someone wouldn't be able to compile old codes with the new compiler (for example variable.name) with a dot_separator wouldn't work in the new compiler. metaQuotes being a pretty good developer would probable get the ex4 support correct (if that's indeed their intend). But I don't see Microsoft (for example) supporting programs for WindowsXp. I have quite a few programs which wouldn't install on windows7 even in Xp/Compatibility_Mode.

metaTrader5 and metaTrader4 offer different features and thats why I believe 4 is still around. mt5 cannot be accepted by the exchanges/sectors which MQ wanted to enter with the features of mt4 included. I don't believe that mt4 is around because of commercial suicide if abandoned. Should mt4 become un_supported, most would just move_onto mt5.

99.999% of MT4 users don't read this forum. They've got EAs which they've bought from websites for $99, they've got .ex4 indicator files which they've downloaded from sites like ForexFactory, and they've got a few things which their broker offers them. If MT4 doesn't have support for legacy .ex4 files then, when the next version comes down the wire, all this stuff suddenly stops working. Does the user say (a) "oh, that's okay, I'll just go out and try to get new versions of everything I was using" or (b) "WTF? All my MT4 stuff has stopped working. I might as well try out that new platform which my broker has been pushing at me."

(Microsoft don't have a 100% record on backward compatibility, but they go out of their way to try to make it 100%. Most famously, in the move from Win3.1 to Win95 they carefully added support for buggy behaviour in the original Civilization game, allowing it to continue doing things like addressing memory after freeing it.)

 
cyclops993:

99.999% of MT4 users don't read this forum. They've got EAs which they've bought from websites for $99, they've got .ex4 indicator files which they've downloaded from sites like ForexFactory, and they've got a few things which their broker offers them. If MT4 doesn't have support for legacy .ex4 files then, when the next version comes down the wire, all this stuff suddenly stops working. Does the user say (a) "oh, that's okay, I'll just go out and try to get new versions of everything I was using" or (b) "WTF? All my MT4 stuff has stopped working. I might as well try out that new platform which my broker has been pushing at me."

(Microsoft don't have a 100% record on backward compatibility, but they go out of their way to try to make it 100%. Most famously, in the move from Win3.1 to Win95 they carefully added support for buggy behaviour in the original Civilization game, allowing it to continue doing things like addressing memory after freeing it.)

Lets be honest, metaTrader4 is aged and shouldn't even be around. I don't think this happened from metaTrader3->metaTrader4. If metaTrader4 support stops, wouldn't this be the same case for the %99.99? Yea I agree, some would try other platforms, but most would just end_up back to metaTrader because #2: thats what they're familiar with. All just my opinions again :)
 
ubzen:
Lets be honest, metaTrader4 is aged and shouldn't even be around. I don't think this happened from metaTrader3->metaTrader4. If metaTrader4 support stops, wouldn't this be the same case for the %99.99? Yea I agree, some would try other platforms, but most would just end_up back to metaTrader because #2: thats what they're familiar with. All just my opinions again :)

I'm not familiar with MT3 but I assume that code that ran on MT3 also ran on MT4 . . . this is the problem with MT4 --> MT5 and is why the situation is different.

The majority of MT4 users can't code, if their favourite Indicators on MT4 don't exist on MT5 then they won't use MT5.

 
RaptorUK: I'm not familiar with MT3 but I assume that code that ran on MT3 also ran on MT4 . . . this is the problem with MT4 --> MT5 and is why the situation is different. The majority of MT4 users can't code, if their favourite Indicators on MT4 don't exist on MT5 then they won't use MT5.

My researching isn't showing the codes to be compatible. I've looked at some of the actual .mql codes and nadda-no.way they're compatible.

I don't believe any other platform would have their custom_indicator. They'll have a better chance recreating/finding it upon mt5.

I have a feeling that the learning_curve for a new platform, community, price, freebies, popularity would still tilt in metaQuotes favor.

Doesn't mean they wouldn't lose some market share due_to differences between mt4 vs mt5.

Best compromise is probably two versions of metaTrader (mt4_like) & (mt5_like). Or get exchanges to accept mt6 (with everything).

That way the risk to losing her core_users because they're Pissed_Off would be minimized.

 
Some points about the Community. I was careful to say Online_Community instead of just mqlx.com. These online community does include other popular forums and the code_base. The non_coders are getting their custom_free_tools from somewhere online | the_code_base. The developers of these tool use these same online_resources extensively. As a developer, I can appreciate the complexity of trying to maintain legacy codes with newer_versions.
Reason: