PowerBasic Forum down (20.07.2013)

[Patrice Terrier]

For those not aware of it, Frederic Laboureur is the name of the french developper who write PureBasic.

Nothing related to PowerBasic of course, and the two languages couldn't be compared (my opinion).

I was thinking more to the comeback of THIII .

...

[José Roca]

John, looks like you are worried by the possibility that PowerBASIC has a future. Please, stop trolling.

[Aslan Babakhanov]

I collected the HISTORY since v.10. For Jim Dunn and others.
May be not complete, but at least some information..

10.4
- Potential miscalculation of XPRINT PREVIEW physical margins.
- IDE: Incorrect display of some international characters.
- Improved scaling of Print Preview text.
- Inaccuracy resizing Graphic controls.
- IDE: Substantial enhancement of Debug Step speed.
- MSGBOX error in a program with no other unicode variables, functions...
- Array Scan errors with certain wide unicode strings.
- Array Sort errors with certain wide unicode strings.
- Errors referencing a wide function using type id's (Call Function$$)
- Errors inheriting the IPowerTime interface.
- IDE: Side effects from certain debugging commands.
- FONT END generates no errors.
- Improvements to XPRINT with a page range.
- IDE: Incorrect display of some international characters.
- IDE: Improvement in debugger stepping.
- UNWRAP$ error with certain short strings.
- Improved error codes for duplicate CallBack name definition.
- BUILD$() function errors using wide unicode parameters.
- TXT.WINDOW() now auto-closes any existing text window.
- Calculation errors using SELECT CASE CONST$$.
- Improved signature validation of linked procedures.
- Occasional GPF with RESUME use in an extracted (unreferenced) procedure.


10.3

- Thread functions/methods defaulted to an incorrect FPU control word.
- Minor changes to the help file
- IDE syntax coloring was disabled when "Use Syntax Color when printing" was not selected and "No Case Change" was selected. This has been fixed.

- Errors with DISPLAY BROWSE in Unicode mode are fixed.
- XPRINT PREVIEW with escapement could be clipped in error and did not allow overwrite of printed items. Both of these issues have been resolved.
- DIR$() now masks system errors.
- DEBUG alignment errors with multi-line macros have been fixed.
- GRAPHIC GET LOC and GRAPHIC SET LOC now only support Graphic Windows. For Graphic Controls use CONTROL GET LOC and CONTROL SET LOC.

- Fixed errors using a non-object parameter with a Thread Object.
- Errors using some forms of Line Input# have been resolved.

10.2

- Errors calculating reverse operators in DWord expressions have been fixed.
- Resolved an error with PowerTime.TimeString for one hour after noon and midnight.

- IPowerTime.NewTime incorrect errors have been fixed.
- Incorrect parentheses counting of CODEPTR as a parameter have been resolved.

- ASMDATA DW now accepts signed values.
- Updated Keyword lists in the IDE.
- Minor changes in the help file.

- Fixed certain compound object reference errors in an SLL.
- Errors executing IDISPINFO methods have been resolved.
- Errors compiling certain rare quad integer literals no longer occur.

- Resolved an issue with GRAPHIC SET VIEW incorrectly setting vertical coordinates in certain instances.
- Errors executing VARIANT$() and VARIANT$$() have been fixed.
- Record length errors with certain random access files have been fixed.
- Fixed an issue where certain uses of EVENT SOURCE caused a compile time error.
- SETATTR no longer causes a GPF when a disk error occurs.
- Certain international characters where not displaying correctly in the IDE have been fixed.
- IDE Close All Files issue on Windows 98 have been resolved.

10.1

- CLIP$ has been revised to remove restrictions.
- Code alignment updates.
- COMMON DIALOGS (DISPLAY) have corrected Ansi/Unicode access.
- IDE Close All Files now correctly closes all tabs/files.

- Certain uses of Graphic Set Pixel required Graphic Redraw to display have been resolved.
- Errors with compiling certain CONTROL ADD "CustomControl" statements have been fixed.

- Mask variable assignment expressions may contain any combination of LongInt and DWord values without error. Operators may include +, -, AND, OR, XOR.
- There is a dramatic improvement of execution speed in many/most DWord expressions.

- ARRAY SCAN errors using WSTRINGZ strings is now fixed.
- Resolved an issue using INPUT #n, x$$ with CHR=WIDE.
- GETATTR errors on Win95/Win98 or using #OPTION ANSIAPI are fixed.
- PATHSCAN$ errors on Win95/Win98 or using #OPTION ANSIAPI are fixed.
- Errors with certain forms of LINE INPUT# are now fixed.
- GETATTR() now generates the correct error value.
- ISFOLDER errors on Win95/Win98 or using #OPTION ANSIAPI have been fixed.
- Fixed a compilation error using CODEPTR() in certain boolean expressions.
- An error with #COMPILER and minor revisions included has been fixed.
- Fixed an issue with Breakpoint errors when debugging a Prefix Block.
- DIALOG NEW PIXES now positions the dialog correctly when the dialog is larger than the screen and no a x and y location is specified.
- ASM LABEL: can now use either ' or ; to add a remark.

- LISTVIEW SORT errors when used with the NUMERIC option.
- TYPE SET statement has revised documentation.
- PowerTime.DateDiff now reports invalid parameters through OBJRESULT.
- PowerTime.TimeDiff now reports invalid parameters through OBJRESULT.
- PowerTime.NewDate now reports invalid parameters through OBJRESULT.
- PowerTime.NewTime now reports invalid parameters through OBJRESULT.
- Compiler error code 639 added: TYPE variable expected


10.0

New Statements and Functions:
- VARIANT$(BYTE, VrntVar) function returns the contents of a Variant as a ANSI byte
  string. This result can be assigned to an ANSI string variable or a User-Defined Type.

Changes to existing Statements and Functions:
- Improved #RESOURCE error checking at compilation.
- FOR/EACH variables are now supported in any scope.
- Fixed errors when calling DIALOG GET CLIENT or DIALOG GET CLIENT with a non-existent
  dialog handle.
- Resolved an issue where LISTVIEW SET HEADER displayed text incorrectly.
- EOF() function now recognizes a soft eof character.
- SLL errors involving the use of an alias name have been fixed.
- Side effects from GRAPHIC COLOR under unusual circumstances has been resolved.
- Errors with PowerTime.DayOfWeekString and PowerTime.DateStringLong in certain locales
  has been fixed.
- PowerTime.TimeDiff has been updated to return the number of Days differnt between
  the two PowerTime objects.
- PowerTime.DateStringLong no longer generates a leading space in some locales.
- Fixed and issue with PowerTime.DateStringLong generating a trailing nul character.
- Fixed issues with LPRINT ATTACH not working correctly in all situations.
- Resolved errors linking Dispatch and private Classes in an SLL.
- Improved Graphic printing with Clear Type enabled.
- Fixed an issue with ClipBoard Get Text and ClipBoard Set Text on Windows 98.
- Syntax erros with GRAPHIC STRETCH PAGE and XPRINT STRETCH PAGE have been corrected.
- Errors with GRAPHIC INPUT using an array variable as a parameter have been fixed.
- Errors when using XPRINT POS(), XPRINT SPC(), XPRINT TAB(), and XPRINT (, column)
  as the first statement on a page have been corrected.
- Resolved an error that caused CONTROL SET FOCUS to be ineffective when called from
  wm_initdialog.
- VARIANTVT() errors with object parameters have been resolved.
- Corrected errors with USING$() with certain unusual string mask parameters.
- VARIANT$ and VARIANT$$ used to return strings based on the contents of the variant.
  VARIANT$ now assumes the contents of the variant is a wide Unicode string and converts
  it to a ANSI string. VARIANT$$ assumes the contents of the variant is a wide Unicode
  string and returns the contents directly as a wide Unicode string.
  VARIANT$(BYTE, VrntVar) always returns the contents as an ANSI byte string. This
  result can be assigned to an ANSI string variable or a User-Defined Type.
- ASMDATA:DW did now supports the full data size.
- Improved #RESOURCE error checking at compilation.
- Fixed errors with PARSE to wide unicode strings.
- CSET/LSET/RSET errors with wide fixed strings have been resolved.
- Fixed Array Scan errors with UDT array and collate option.
- LISTBOX/COMBOBOX FIND errors on Win98 have been fixed.
- Fixed an error using DIR$() on Win98.

Additional changes:
- Compiler error code 461 changed: INSTANCE arrays must be declared. INSTANCE arrays
  must be declared before any CLASS code.
- Suppress name clashes with import functions in an SLL.

------------------------------------------------------------------------------
                                     IDE
                                     ===
- Fixed an issue with fonts appearing clipped if Windows font smoothing (e.g., ClearType)
  was enabled.
- In "single instance" mode, trying to launch a second IDE will cause its
  command line (if any) to be passed to the original IDE. So, the original IDE
  will load any files that would have been loaded in a second instance.
- Variable Watcher properly restores the sizes of its listview columns.
- The display bounds checker fully supports the use of multiple monitors.
  The IDE will re-open on the appropriate display.
- Fixed Code Finder handling of PROPERTY SET. The Type information for PROPERTY
  now distinguishes between PROPERTY GET and PROPERTY SET.
- The colors of the Output Window match better with Windows Themes.
- Saving backup files with timestamps uses the correct timestamp again.
- Double-click in Variable Watcher is ignored for empty rows. Evaluate Variable is
  enabled only if there is a symbol name at the caret.
- The last specified file path is selected as the default path regardless of how the
  last file was loaded.
- Backups may now be done to a specified directory. The IDE will attempt to create
  the directory if it does not already exist. The default setting is ".\", the path
  of the saved file.
- Added Shift+Delete as "Cut" key.
- Ctrl+F4 added as "Close File" key.
- Copying to the clipboard is enabled while debugging, if appropriate.
- Insert File is disabled while debugging.
- Files loaded by virtue of being stepped into while debugging are adjusted to match
  their names on disk, to avoid undesirable case conversion.
- PBForms can be launched from PBEdit even if no files are loaded.
- Adjustments to Options / Fonts handling for proper font listing under Win9x.
- Compiler options now has a checkbox to specify if a .PBR file should be created
  when compiling an .RC file.
- File options now has a checkbox to select which files are included when saving
  a project.
- General options now has a check to select whether to display the IDE status bar.

------------------------------------------------------------------------------
                                     COM Browser
                                     ===========
- Tools | Options dialog box now disables the OK button and allows the Cancel button to
  be used when the Interface Prefix textbox is null.
- Added a fix for interfaces that have an incorrect inheritance.

[Mike Stefanik]

I think we're having two different conversations here, with folks talking (or writing, as the case may be) across one another. What I find to be most interesting are those who claim simply that they've purchased a product, and as long as that product continues to work then why should anyone have an issue with "free" resources like a website. Frankly, I see that as an outmoded perspective on the business of software.

The software business is a service business.

The software itself is an important part of that business, of course, but it is not the only part. In the 21st century, that means a working website were customers can download updates, obtain information about your products and even order them. It also means providing a means for you to communicate with your customers both privately and publically, and if they choose, amongst themselves. That is not a "freebie", it is an essential part of providing excellent service for the people who depend on your software to get their work (or play) done. If you think of a website and forum as some kind of vestigial tail on a software business, then simply stated, your thinking is wrong.

As to this specific situation, I'll just say as someone who's been involved with the coding, the site administration and some of the business aspects of software for a lot of years now, there is absolutely no situation where I would consider it acceptable for our site (or any business site) to go dark for days on end. At the very best, it demonstrates a complete lack of planning and technical acumen by the staff responsible. It is indeed possible to move offices, whether it is down the street or across the country, with minimal or no downtime, even if your primary servers are in-house. But of course, you need to plan and coordinate that move, and have contingency plans in place if something does go wrong. That's also part of being a responsible business in service to your customers, and frankly, just good business sense.

If the reason they've been down for so long was because of a DDoS attack or something of that nature, then I would say that it's reasonable to cut them some slack. But this seems to be an unforced error on their part, and I think some criticism is justified, particularly since this is not the first time that this sort of thing has happened. While some of you who have been customers for many years are understandably giving them the benefit of the doubt, you also need to accept that you are not the future of the company. The future is new customers, new developers who learn to appreciate the same things you do about the language. And if you were that new developer who heard good things about PowerBasic and found the website down for days without explanation, what would you think? Would you risk your reputation, and even your career, recommending a solution to your company when the software vendor cannot even maintain basic web services for their own customers? Where you couldn't even order the product online if you wanted to?

For those of you who disagree with the criticism, I would point out that this is not the worst thing that can happen to a business. In many cases, people are most critical when they have a vested interest in the outcome. They are vocal in their opposition or express their disappointment because they want the company to succeed and continue to be able to provide the product and services that they have grown to depend on. When a business needs to be truly concerned is not when customers are unhappy and complaining; it's when no one is complaining because they're past caring and have moved on.

[John Spikowski]

John, looks like you are worried by the possibility that PowerBASIC has a future. Please, stop trolling.

Bob is gone. I'm (and everyone else) no longer required by law to only speak about PowerBASIC in a reverent way and be thankful Bob took a moment of his time to respond to our silly questions. I thought we were having a debate about the future of PowerBASIC and didn't realize the full scope of the rules. I guess I tilted out and now the forum Troll.

[Patrice Terrier]

Mike,

I fully agree with all your comments.

Even if some are groupies, we are first customers.

As i said many years ago to Mr. Zale:  you made us duke, but who made you king

...

[Michael Mattias]

I have working development tools create 32-bit applications: PB/CC and PB/Win.

I have at least peer support from multiple sources.

I am currently planning on needing these things for a couple years.

As long as a32-bit applications live a couple years more, I'm happy as a camel on Wednesday.

Those who have to think ahead more than a couple of years may not be so happy.

MCM

[John Spikowski]

Welcome Michael!

Your take on things is refreshing and informative with a touch of wit and honestly that is your signature. You may need to oil that armor and sharpen the sword as the troops are restless and hungry and taking no prisoners.

[José Roca]

John, looks like you are worried by the possibility that PowerBASIC has a future. Please, stop trolling.

Bob is gone. I'm (and everyone else) no longer required by law to only speak about PowerBASIC in a reverent way and be thankful Bob took a moment of his time to respond to our silly questions. I thought we were having a debate about the future of PowerBASIC and didn't realize the full scope of the rules. I guess I tilted out and now the forum Troll.

Some of us are having a debate about the future of PowerBASIC. You're just trying to discourage everybody. Apparently, Bob's death is not enough to you, you would also like to destroy his legacy. Which interest can you have in the future of PowerBASIC if you have said that you don't use it?

[John Spikowski]

You're just trying to discourage everybody. Apparently, Bob's death is not enough to you, you would also like to destroy his legacy.

Bob thought he owned the BASIC market and no one was as smart as him to build a BASIC compiler worth using. He was greedy with an ego as large he was physically. I think it will come out in the end that Bob's homemade assembler is impossible to use by anyone other than him. Bob proved only one thing, you can take it with you.

While PB struggles to find itself, I released ScriptBasic 2.2 for beta on multiple Linux platforms and Windows and working with Charles on enhancing the Windows version of SB. I also ported two BASIC translators and two interpreters to Android Linux and compiling native on the device with gcc C/C++. Does that sound like all I have to do with my spare time is be a troll on the JRS forum? 

I will NOT comment again on anything related to PowerBASIC or it's author. There are more than enough past users that can pick up where I left off.

[Mike Stefanik]

José, I think the word you're looking for there is schadenfreude and I'd agree, there does seem to be some of that. However, I think for most folks it's simply the discomfort of feeling left in the dark with no clear understanding of what's happening now, along with the lack of any clearly articulated plans for the future.

Note I'm not talking about pushing vaporware here, I'm talking about a roadmap that defines some specific plans for where things are going with the language and a general timeframe in which they would like to accomplish them (with the acknowledgement that nothing is set in stone). I'd also suggest that they publically discuss the steps that they're going to take to ensure that they don't completely fall off the grid again. If they did this, I suspect it would put an end to much of the hand-wringing that has been going on.

[Bernhard Fomm]

The forum is offline. For a day goes something like this once through as a worst case.
But five days? What is this? What are the reasons?

The developers are worried if you were dependent on the forum and the support.

[Guy Dombrowski]

Anybody know where Steve Rossel went ?

Support quality was never the same when he left.

[Colin Glenn]

I have working development tools create 32-bit applications: PB/CC and PB/Win.

I have at least peer support from multiple sources.

I am currently planning on needing these things for a couple years.

As long as a32-bit applications live a couple years more, I'm happy as a camel on Wednesday.

Those who have to think ahead more than a couple of years may not be so happy.

MCM

64 bit systems are just 32 bit systems that are twice as wide. Anyone who thinks 64 is superior to 32 is just plain in ignorance of how machine architecture works. Yes, it is speedier. Yes, it does work with a wider data path. But, for the most part, 32 bit applications will probably run just as fast as 64 bit applications due to architecture.

CAPTCHA does not work for me and my eyesight, ...

[Patrice Terrier]

Guy--

Steve Rossel had the opportunity to retire earlier to spend more time with his kids and he could not let pass it up.

...