What's new in MapInfo 9.5

[Naz Ali]

The conference is almost over! Thought I'd give a highlight of the new
features of MI 9.5 releasing in June.

EASE OF USE
- few menu items added to the right-click, such as, view entire active
layer and select all from layer selection
- the active layer is bolded in the layer control
- there is grid control for the drop-down menus for colours, symbols,
lines and patterns
- in the new map window, you can choose layers by adding/removing
- in the layer control, you can drag the layers up and down rather
than clicking up/down many times
- the Options>Preferences now has a WinXP look

CARTOGRAPHIC FEATURES
- automatic labelling around points
- now allows curved label overhangs
Custom Symbols:
- supports jpg, tiff and gif
- these symbols can be in a heirarchy of folders
- the location of custome symbols can be configured
- there are, i believ, 226 new custom symbols
Vector Translucency (targeted for November release)
- translucency for objects, selections, thematic maps and even labels.

NEW DATA EDITING TOOLS
- 40 new tools for data creation and editing
- MapCAD toolbar, such as, calculate angle, split regions, create
parallel, select objects by style, etc. This can be customized and
comes with full user documentation.

EXTENDED DATA ACCESS
- Support for MS SQL,2008,spatial data format via open universal
option
- some new and updated projections
- Programming with .NET
- design your GUI in .NET and run within MI Pro
- MapBasic Functions, such as, RasterTableInfo(), GridTableInfo(),
ControlPointInfo()

ENHANCED LICENSING & DEPLOYMENT
- share MI customizable files with other users

Please add what I may have missed.

Kind regards, Naz

[Gajendra Salunkhe]

Hi

Where can we find the documents on MapInfo 9.5, I am more interested its working with .NET technology.

Regards

[Eric_Bl...@mapinfo.com]

A few quick comments.

The layers(s) (could be more than one)  are bolded whenever the selection is from the table the layer is part of. There is no "active" layer in Professional.
You have been able to drag layers up and down in layer control since version 4 released in 1995. This is not a new feature.
All the dialogs use an XP style look if you have that option turned on in your Windows settings.
Custom symbols now support any raster type that we support as raster data. Notable omissions below would be .png but there are others, of course including custom handlers that you or some third party develops.  Note that some multi-file raster types might not be listed in the UI for adding a symbol. This is because we don't have the logic that knows that some random raster of extension .XXX  also uses files .YYY, etc. However, if you put all the required files in the custom folder, it should all work. We are rendering these symbols with the same engine that renders them as a raster table.

The programming with .NET is probably of some interest to this group. Some of the more advanced users have already done this so you might be thinking that you don't need this. These are the highlights:

• A .NET static method can be called as simply as calling a DLL function.  However, better type checking can be done with .NET. There is no setup work you need to do.
• The .NET CLR is hosted by MapInfo Professional. This means we can catch exceptions and get events when things are happening in that world. Previously, we would crash or strange window behavior could result because we had no knowledge of what was happening in  .NET. The stability of the feature is therefore much greater.
• You don't have to put assemblies in the GAC (or forbid, the Professional program directory) to have your assemblies load with stability. An assembly can just be in the same folder as the .MBX.
• Callbacks are supported and have been enhanced to handle multiples.
• The Named Views sample has been rewritten as one example of using .NET.

Eric Blasenheim
Chief Product Architect
Pitney Bowes MapInfo Corporation


Mail List:grbounce-yvy1equaaaajbprysysrydkk7vpghp_9=mail_list=mapin...@googlegroups.com

From:	Naz Ali <naz...@cbre.com> on 05/23/2008 10:31 AM MST
To:	MapInfo-L <mapi...@googlegroups.com>
cc:
Subject:	[MI-L] What's new in MapInfo 9.5

[Gregory...@mapinfo.com]

We will be launching a web page for the 9.5 release in about a week or so.  I'll post the url here once it is live.  

	Greg Donahue      Senior Marketing Manager
	Pitney Bowes MapInfo One Global View |Troy, NY 12180

"Gajendra Salunkhe" <gsal...@gmail.com> Sent by: mapi...@googlegroups.com 05/26/2008 05:27 AM Please respond to mapi...@googlegroups.com

To mapi...@googlegroups.com cc Subject [MI-L] Re: What's new in MapInfo 9.5

[Glen]

sorry guys but I am gonna vent!

"The programming with .NET is probably of some interest to this group.
Some
of the more advanced users have already done this so you might be
thinking
that you don't need this. These are the highlights:"

yup with no help or support from MapInfo ;-) and documentation that
has not been updated in 7 years.

How about less sales talk and more "REAL" documentation. maybe some
code examples on a current .net language?

Glen

[scisoft]

Eric, thanks for the pre-release information. It sounds very positive.

Can you amplify slightly on what you’ve told us recently? I know it will all come out in the wash, but it would be nice to have something more explicit.

The programming with .NET is probably of some interest to this group. Some of the more advanced users have already done this so you might be thinking that you don't need this. These are the highlights:

You jest, surely? I have gained the impression since the demise of Project Grande that MapInfo Corporation has studiously ignored assisting anyone to develop even the most trivial of solutions to add into MapInfo. Perhaps that was because it was so brittle an interface.

·       A .NET static method can be called as simply as calling a DLL function.  However, better type checking can be done with .NET. There is no setup work you need to do.

So, I assume that there is a new MapBasic 9.5 that supports some additional syntax, to interop with the .NET assemblies. How much change is there from MB 9.0 to 9.5? How does a MBX (? still the same package, I assume) recognize that it is hosted on 9.5 and has the capacity to recognize a .NET assembly (DLL)? Are there “environment variables” that allow that?

And will MB 9.5 with .NET functionality be a free download, so that those who don’t own MapInfo 9.5 (and may not want to) can develop solutions for it?

·       The .NET CLR is hosted by MapInfo Professional. This means we can catch exceptions and get events when things are happening in that world. Previously, we would crash or strange window behavior could result because we had no knowledge of what was happening in  .NET. The stability of the feature is therefore much greater.

Following on from the previous question, if a MBX and its accompanying .NET assembly (DLL) makes itself known to MapInfo 9.5 (via some install process – hopefully a standardized deployment recipe will be forced on developers) then I assume that there are fool-proof methods to check for full installation of the .NET v2.0 runtime. What are they?

While not understanding what you mean by “hosting” the CLR (I assume this is some sort of COM interop that manages 2-way communication between the Win32 MapInfo core, and the .NET CLR runtimes), what .NET functionality cannot be used in a .NET assembly / “add-in”?

·       You don't have to put assemblies in the GAC (or forbid, the Professional program directory) to have your assemblies load with stability. An assembly can just be in the same folder as the .MBX.

So, the third-party / foreign / user-written DLLs (assemblies) may be placed in a typical SxS configuration (and I assume use the usual XML .config files) in an “add-in” folder location, that is made known to MapInfo during its installation?

·       Callbacks are supported and have been enhanced to handle multiples.

Can you explain that more fully? And do you have templates that will be supplied for both Visual Studio 2005 and Visual Studio 2008 that allow the authoring of callbacks that are not brittle and error-prone?

·       The Named Views sample has been rewritten as one example of using .NET.

Will there be some high-calibre examples, too?

A couple of other questions:

1. I’m assuming you’re targeting .NET v2.0 – yes? Or have you gone for the much more capable .NET v3.5? With 3.5 SP1 and VS2008 SP1 the .NET Framework Client Profile is a very compact install.
2. Since you’re able to deal with SQL Server 2008 “spatial” and Oracle 11g, what is the support for LINQ to SQL?
3. Is there any “assistance” through the MapInfo 9.5+ API for enhanced graphic support (WPF, GDI+, CUDA)?

Again, thanks for the clarification.

---

Ian Thomas
GeoSciSoft  - Perth, Australia

[scisoft]

Eric

I forgot an important question.

Will you be providing Intellisense (via the MapInfo interops, of course) for Visual Studio?

And while I’m at it, those programming in .NET will have a Rolls Royce UI, but what about the creation of the MBX side of the equation? Is there some facility for a better UI editor for poor old MapBasic?

---

IL Thomas
GeoSciSoft  - Perth, Australia

[Mags]

Hi Eric

My question is whether there was any announcement related a Mapx Mobile for Windows CF 2.0 !

 

Thank you !

[Eric_Bl...@mapinfo.com]

Glad to see that this is generating some interest. By the way, I refrained from answering the previous questions "in line" because I did not want to assume that everyone's mail system would show colors and without that those type replies tend to be unreadable. I don't know if that is an issue for most people anymore.

Anyway, most of Ian's questions are better answered with just some information. Let me know if you need more.

• Ian is absolutely correct about our lack of support for writing add ins in .NET. This is because the product did not support it. Truthfully! There were too many ways to make it misbehave or crash.
• If you would like to read up more on what hosting the CLR means, here's one link. http://msdn.microsoft.com/en-us/library/9x0wh2z3(VS.80).aspx but basically we are manually loading the CLR (Common Language Runtime) and have greater control over what happens.
• My comment about some people saying they did not need support for .NET is a direct result of some folks telling us just that. THEY did it so it was fine. I was not jesting. For most users I believe what we have done in 9.5 will have real value and seriously improve the capabilities of MapBasic programs.
• MapBasic 9.5 will be a free download like 9.0. As most of you know, MBX's have a version number that is just the version of the compiler and therefore is sometimes higher than it needs to be. We have had this discussion before. However, when using .NET in an MBX this version restriction is real. The earlier versions of Professional would not know how to do this.
• There is an entirely new chapter in the MapBasic documentation devoted to .NET work. The Professional team is always interested in your comments on documentation.
• There is a new extension to "Declare" which takes a method name and a class name. The method must be static as MapBasic cannot instantiate classes. However, this can just mean that the static method is a way "in" to the .NET world.
• The new Declare syntax is the only real change in the .MBX. The rest of the big work is all in Professional which is why a new .NET connected MBX cannot be used in an earlier version.
• There is no registration needed for assemblies. The goal was to make using a method in .NET as simple as a DLL call with, as it turns out, much better run time type checking thanks to .NET which is very useful for debugging a Declare statement with incorrect parameters. One can use the GAC but one does not have to. One can simply put the assembly in the same folder as the MBX which was fairly typical for a dll.
• The .NET 2.0 Runtime will be installed as part of the Professional installation if it is not already there. If you manually uninstall the framework, MapInfo Professional will no longer load. This is not just because we need it for MapBasic extensions but for our own internal use.  
• The callbacks that I referred to are actually an enhancement to our COM interface exposed in .NET. In versions prior to 9.5, only one callback "listener" could be registered at a time. When the callbacks are used from within the our process space (e.g. MapBasic app that calls into a user defined DLL that interacts with the COM interface), supporting one callback listener is insufficient, so we enhanced Pro to support multiple listeners.  The Integrated Mapping section of the User Guide has been updated with the new COM interface methods that support this enhancement.
• We have defined a .Net helper class that wraps a subset of the COM interface.  This includes wrapping the callbacks as true .Net events.  In order to use the wrapper class in your Visual Studio project you must reference a particular assembly (miadm.dll) in our install folder.  You can find more details about using the wrapper class in the MapBasic User Guide.  Once this assembly is referenced, intellisense will work.  We do not provide Visual Studio templates, so you have to create the assembly reference manually through the Add Reference dialog.
• At this time we have just few samples but that will change. We have the new Named Views, a sample showing use of .NET for integrated mapping and a hello world getting started sample in both C# and VB.NET. We hope this will be expanded.
• There is no way to know what functionality of .NET cannot be used be used. However, as Professional is a single threaded application, doing things with separate threads may be problematic. There are special notes about callbacks and threading in the documentation.
• As stated already, we are using the 2.0 .NET framework which is the foundation of 3.x. Nothing we are doing requires 3.x but you are free to use it. However, if you wish to deploy that application, your installer will be responsible for ensuring any of the 3.x extensions get installed. I will take it as a note that we should look at ways in which 3.x could be used.
• LINQ was mentioned so I think it is appropriate to state clearly that MapBasic is now allowing you to extend Professional with .NET. There is no reason why you should not be able to use LINQ in your extension but nothing like that will change the essence of Professional. So, if you were wondering whether LINQ syntax could USED in Professional, the answer is no.
• I am not sure what was meant by "assistance" through our API for enhanced graphic support. However, we will be using GDI+ in our implementation of translucency and other graphics enhancements hopefully coming by the end of the year.
• There is nothing new in creating dialogs directly in MapBasic. As the list has shown, there are a number of editors and tools that can help with this. We have stated many times that while we can and should look at items like enhancing the size of an MB, enhancing the editor or building tools for UI is something that is outside of our focus and takes away valuable resources from providing more value in the core product. We hope that the support for .NET will encourage people to use that mechanism for building UI wherever possible.

The team will be watching this thread for reactions for sure!

Eric Blasenheim
Chief Product Architect
Pitney Bowes MapInfo Corporation


Mail List:grbounce-yvy1equaaaajbprysysrydkk7vpghp_9=mail_list=mapin...@googlegroups.com

From:	"scisoft" <geosc...@iinet.net.au> on 05/27/2008 09:58 PM ZE8
To:	<mapi...@googlegroups.com>
cc:

Subject:	[MI-L] Re: What's new in MapInfo 9.5

Eric, thanks for the pre-release information. It sounds very positive.
Can you amplify slightly on what you've told us recently? I know it will all come out in the wash, but it would be nice to have something more explicit.

The programming with .NET is probably of some interest to this group. Some of the more advanced users have already done this so you might be thinking that you don't need this. These are the highlights:

You jest, surely? I have gained the impression since the demise of Project Grande that MapInfo Corporation has studiously ignored assisting anyone to develop even the most trivial of solutions to add into MapInfo. Perhaps that was because it was so brittle an interface.

·       A .NET static method can be called as simply as calling a DLL function.  However, better type checking can be done with .NET. There is no setup work you need to do.

So, I assume that there is a new MapBasic 9.5 that supports some additional syntax, to interop with the .NET assemblies. How much change is there from MB 9.0 to 9.5? How does a MBX (? still the same package, I assume) recognize that it is hosted on 9.5 and has the capacity to recognize a .NET assembly (DLL)? Are there "environment variables" that allow that?
And will MB 9.5 with .NET functionality be a free download, so that those who don't own MapInfo 9.5 (and may not want to) can develop solutions for it?

·       The .NET CLR is hosted by MapInfo Professional. This means we can catch exceptions and get events when things are happening in that world. Previously, we would crash or strange window behavior could result because we had no knowledge of what was happening in  .NET. The stability of the feature is therefore much greater.
Following on from the previous question, if a MBX and its accompanying .NET assembly (DLL) makes itself known to MapInfo 9.5 (via some install process - hopefully a standardized deployment recipe will be forced on developers) then I assume that there are fool-proof methods to check for full installation of the .NET v2.0 runtime. What are they?

While not understanding what you mean by "hosting" the CLR (I assume this is some sort of COM interop that manages 2-way communication between the Win32 MapInfo core, and the .NET CLR runtimes), what .NET functionality cannot be used in a .NET assembly / "add-in"?

·       You don't have to put assemblies in the GAC (or forbid, the Professional program directory) to have your assemblies load with stability. An assembly can just be in the same folder as the .MBX.

So, the third-party / foreign / user-written DLLs (assemblies) may be placed in a typical SxS configuration (and I assume use the usual XML .config files) in an "add-in" folder location, that is made known to MapInfo during its installation?

·       Callbacks are supported and have been enhanced to handle multiples.

Can you explain that more fully? And do you have templates that will be supplied for both Visual Studio 2005 and Visual Studio 2008 that allow the authoring of callbacks that are not brittle and error-prone?

·       The Named Views sample has been rewritten as one example of using .NET.

Will there be some high-calibre examples, too?
A couple of other questions:

1.        I'm assuming you're targeting .NET v2.0 - yes? Or have you gone for the much more capable .NET v3.5? With 3.5 SP1 and VS2008 SP1 the .NET Framework Client Profile is a very compact install.
2.        Since you're able to deal with SQL Server 2008 "spatial" and Oracle 11g, what is the support for LINQ to SQL?
3.        Is there any "assistance" through the MapInfo 9.5+ API for enhanced graphic support (WPF, GDI+, CUDA)?

Again, thanks for the clarification.

---

Ian Thomas
GeoSciSoft  - Perth, Australia

[Glen]

"Professional is a single threaded application, doing things
with separate threads may be problematic. There are special notes
about
callbacks and threading in the documentation."

Can you give more detail on this statement?
using a single thread would be a speed / performance problem

.Net 2.0 is already outdated, I agree with Ian that 3.5 SP1 would be
they way to go.

[Uffe Kousgaard]

If a single thread is a speed / performance problem for you, MapInfo Pro
probably isn't the right platform at all. MapInfo Pro is a disk-intensive
application, so multiple threads rarely helps a lot. Unless you split the
data between multiple physical disks.

Also, mapbasic is far from offering threaded programming.

Regards
Uffe Kousgaard

----- Original Message -----
From: "Glen" <Glen_...@ossman-cg.net>
To: "MapInfo-L" <mapi...@googlegroups.com>
Sent: Thursday, May 29, 2008 2:44 PM
Subject: [MI-L] Re: What's new in MapInfo 9.5

>

[Eric_Bl...@mapinfo.com]

Glen and all,
3.X .NET framework is really an extension on the 2.0 framework. Many of the core assemblies are versioned 2.0 so I see basing what we have done on 2.0 as absolutely the right decision. Forcing 3.x would offer customers nothing.

Eric Blasenheim
Chief Product Architect
Pitney Bowes MapInfo Corporation


Mail List:grbounce-yvy1equaaaajbprysysrydkk7vpghp_9=mail_list=mapin...@googlegroups.com

From:	Glen <Glen_...@ossman-cg.net> on 05/29/2008 05:44 AM MST
To:	MapInfo-L <mapi...@googlegroups.com>
cc:

Subject:	[MI-L] Re: What's new in MapInfo 9.5

"Professional is a single threaded application, doing things
with separate threads may be problematic. There are special notes
about
callbacks and threading in the documentation."

Can you give more detail on this statement?

using a single thread would be a speed / performance problem

[Glen]

Hi Eric

good spin ;-) so microsoft just wanted to develop 3.0 and 3.5 because
they weren't busy? they really didn't want developers to use it?
all thoses new features and UPDATES aren't really needed? I could go
on but I don't want to seem mean

"The .NET Framework provides the necessary infrastructure to write
multithreaded applications. A thread represents a part of a program
which can be run independently. It is also the basic unit to which an
operating system allocates processor time.Having multiple threads in
this way can improve the performance of the application in a big way.
An efficient way to manage multiple threads in the .NET Framework is
to use the ThreadPool class."by S. Vikram

http://www.developer.com/net/cplus/article.php/2202491 << some
background on threads


Tools like PlanetEV and Actix would benefit greatly from correctly
implemented multi threading support from Mapinfo. (2 tools I use all
the time)

I could look up the word "OUTDATED" for you if you would like?

G

[scisoft]

Whoah - slow down a bit there!

For those on the margins (of the direction this discussion has headed), it
should be explained that "threads" and "threading" is just a programming
technique that allows a program to do more than one thing at once.

Typically, it's of use when there is a disk-intensive task that might be
done in the background, while the user is able to do "other things" - for
example, in the UI (user interface) while that search / disk activity is
happening.

A similar background task is the rendering of complex graphics (or,
computationally-intense calculations, or rendering of vector graphics).

With multi-processor / multi-core computers, a "variation" of the separate
"thread" advantage is the ability to lay off different jobs to a different
CPU or core. Parallel processing is a related term.

Yet another way to get the calculation-intensive stuff done efficiently is
to use the graphics processor (GPU) on the graphics card to do the rendering
(or, other functions). One such system for parallel programming on the GPU
is the CUDA system on (some, newer) NVidia graphics cards.

Also, just buying a computer with a faster CPU and say 2 or 4 cores, and
more RAM, and a faster hard disk, just doesn't give you the pay-off that it
used to 5 or 10 years ago. The program that is running on that hardware
needs to be able to do things that weren't so important a decade ago.

None of these things requires .NET. A 'native' Win32 program can be coded
with things like "thread pools". (Actually, a Windows application can
consist of more than one process, and a process can consist of more than one
thread. All the above is a gross simplification.)

So, MapInfo Professional and MapBasic *could* be made to (could have been
allowed to [1]) handle thread pools, do parallel processing, etc.

But .NET makes programming of threads, parallel processing, and even use of
CUDA scripting / language easier than it is with C++ running on 'native'
Win32.

[1] That's why the abandonment of Project Grande - when the MapXtreme work
had been progressing, and continues to develop - is such a shame, and a
missed opportunity.

Ian Thomas
GeoSciSoft - Perth, Australia

[Glen]

Ian

Nice job in explaining that

G

[MidNight Mapper]

Yes, I agree. Depending on a technical vision co-opted by marketing that
was delayed by literally years through the Project Grande illusion was a
investment lost for developers and MapInfo as well IMHO. Maybe what we are
seening is technical vision returning? But we need a bri9ghter lamp out of
the darkness...

[Eric_Bl...@mapinfo.com]

I will keep my comments pertinent to "what's new in MapInfo 9.5" and make sure we are communicating.

• What we have done in Pro 9.5 would not take advantage of anything in 3.x .NET and therefore there was no reason to have built on top of it.
• You are still free to use anything in 3.x framework in what you add.
• 9.5 just provides an easy connection to get into .NET code from MapBasic.
• 9.5 is NOT a .NET product. That is, the overwhelming amount of code is native windows application code.
• Expecting a connector into .NET to suddenly make the product use multiple threads is, in my opinion, not reasonable and certainly never stated nor implied by me.
• I think it would be appropriate to start another "thread" (pun intended) on performance issues.

[scisoft]

Glen, I think that's a bit harsh.

In my opinion, .NET 3.5 offers some definite incremental improvements over
2.0, but they're marginal unless MapInfo can take advantage of them.

Two areas that cannot be used in MI/MB as it is currently written are the
CLR Addins pipeline model (which gives huge and safe gains in the
development of flexible / extensible apps, useful in some circumstances),
and the use of LINQ (which may or may not be useful for getting spatial and
non-spatial data from enterprise databases, and consuming it in the GIS).

For example, if written for .NET 3.5 a utility like FME (or MapInfo itself)
could have many hundreds of data conversion / translation plugins or addins.
Using the CLR addin model, a very consistent model that allows discovery,
loading and unloading, etc can be developed. Instead of a new version of
FME, a new format could be added by delivering a single DLL. Or, a buggy
translator could be fixed by delivering a new version of a small DLL
component, in the same way.

But, as FME shows, there are other ways of doing data conversion (and, of
course there are other plugin models that don't depend on .NET).

Other advances in 3.0 and 3.5 are language things that could be used in your
external .NET application, but aren't affected by MapInfo's dependence on
the 2.0 version. You can always install 3.5 as a part of your application
("integrated mapping with .NET"), if some extra features are helpful to your
coding. Unless there's more interactivity (non-technical expression) between
MI and .NET then as Eric said, there's not much sense installing the
incremental extras for 3.5 (?adding another 20Mb, probably).

Ian Thomas
GeoSciSoft - Perth, Australia

-----Original Message-----
From: mapi...@googlegroups.com [mailto:mapi...@googlegroups.com] On

Behalf Of Glen
Sent: Friday, May 30, 2008 1:19 AM
To: MapInfo-L
Subject: [MI-L] Re: What's new in MapInfo 9.5

[Glen]

My last statements was harsh.

But having developed VB6 applications that needed to be updated
to .Net supporting Vista I have a special “LOVE” for MapInfo ;-)

G

[Eric_Bl...@mapinfo.com]

One last point. A few other folks on the Professional team have reminded me that another decision point in basing our .NET integration on the 2.0 framework was the fact that we still support Windows 2000 and .NET 3.x is not supported on that OS version.

[Chandler Coleman]

Windows 2000.. I am somewhat confused.  Please let's see how many people are using that OS.  Please reply if you, or if you know someone, that still has that OS up and running on a viable computer.  In my opinion supporting Windows 2000 does not necessarily make good business sense. It really sounds that Pitney Bowes is not using its best judgment here.  They could have offered something useful and choose not too.

Chandler Coleman

 

----- Original Message ----
From: "Eric_Bl...@mapinfo.com" <Eric_Bl...@mapinfo.com>
To: mapi...@googlegroups.com
Sent: Sunday, June 1, 2008 8:32:16 AM
Subject: [MI-L] Re: What's new in MapInfo 9.5

[Eric_Bl...@mapinfo.com]

I will repeat, hopefully for the last time, that building our .NET integration on 3.x, which would just have effectively meant that we would have forced that install instead of 2.0, would have really offered nothing to you. You are not missing anything because of that decision so picking the lowest version that offers what we need IS the right decision.

Deciding when to drop support for an OS version is always an interesting challenge as you need to find out what customers cannot or will not upgrade and therefore would be closed out from the new version.  When the older version is limiting features that might be available to most, then there is a compelling reason, particularly when the vendor, in this case Microsoft, is dropping support.  I will be surprised if even this more advanced group that follows MapInfo-l does not have Windows 2K users. However, like I said, there is no compelling feature reason to remove support at this time.

Eric Blasenheim
Chief Product Architect
Pitney Bowes MapInfo Corporation

Mail List:grbounce-yvy1equaaaajbprysysrydkk7vpghp_9=mail_list=mapin...@googlegroups.com

From:	Chandler Coleman <chan...@yahoo.com> on 06/01/2008 08:45 AM MST
To:	mapi...@googlegroups.com
cc:

Subject:	[MI-L] Re: What's new in MapInfo 9.5

[Lars I. Nielsen (GisPro)]

I do. And I know several other people that also do.

I also use Windows XP, and 2003, and 2008, when the need requires it,
but my primary work laptop still runs 2000.

It's a common mistake to assume that everyone always uses the latest
version of whatever software is out there, and assume that earlier
versions are crap. This is just a markering gimmick (lie) from (in this
case) Microsoft, nothing more.

Sometimes it makes perfect sense to stay with an older, stabile, and
trusted environment, than to experiment unnecessarily with a new, and
waste tons of time on trivial setup (and security) issues. I've never
reinstalled my OS, and never will, unless forced to do so.

I applaude the decision to stay with Windows 2000 support, cheers Eric !

However, my next laptop will most likely be running VMware, so I can run
any OS I need/want on the same machine.

Best regards / Med venlig hilsen
Lars I. Nielsen
GIS & DB Integrator
GisPro

Chandler Coleman skrev:

[Gentreau]

I can also inform you that I know a major mobile network operator with many
thousands of PC's and a MapInfo corporate licence, which is still running
Windows 200 on most of their technical machines.

I would also echo what Lars just said. Very often there is no compelling
reason for a company to upgrade many hundreds or thousands of machines, with
all the potential compatibility issues which can and do raise their heads
each time this happens. I know of other UK companies which made the jump
from NT4 to XP without ever installing W2000. Those who did go to W2000 are
quite likely to be sticking with it in many cases.

Windows 200 is actually quite a good OS, I have it running on a 300Mhz
Pentium 1 machine with 80MB of RAM, bet you couldn't do that with XP !! That
particular machine has a very simple but repetitive job to do and it does it
fine, sure it takes 10 minutes to reboot but that doesn't happen more than
once a year.

Don't fall for the marketing scams, sorry messages :)

BR
Gentreau.

-----Original Message-----
From: mapi...@googlegroups.com [mailto:mapi...@googlegroups.com] On

[Uffe Kousgaard]

> Please let's see how many people are using that OS.

My webserver is Win 2000 Server. Upgrading a perfectly working server costs
money and time and brings no benefits. So, not upgrading makes good business
sense.

Regards
Uffe Kousgaard

[Glen]

Gents

You have hit my point exactly! When telco X decides to upgrade from
window 2000 to a current operation system
Do you think they will pick XP or Vista? How long will their in-house
software be used before it has to be updated (return on investment)

and when they hire a developer to update their in-house software, how
long should that software be good for?

Should the developer use the most current version of the operation
system, so that he might maximize return on investment for his client?
( length of time the client can use the software without an update)

Something to think about

and one more point.

were NOT talking about users here, we are talking about developers and
what tools they have to use.

Glen

[Richard Greenwood]

On Sun, Jun 1, 2008 at 9:45 AM, Chandler Coleman <chan...@yahoo.com> wrote:
> Windows 2000.. I am somewhat confused. Please let's see how many people are
> using that OS. Please reply if you, or if you know someone, that still has
> that OS up and running on a viable computer. In my opinion supporting
> Windows 2000 does not necessarily make good business sense. It really sounds
> that Pitney Bowes is not using its best judgment here. They could have
> offered something useful and choose not too.
> Chandler Coleman

I regularly run MapInfo on Windows 2000 and I commend PB for
continuing to support the platform.

--
Richard Greenwood
richard....@gmail.com
www.greenwoodmap.com

[Mats Elfström]

Hi!
About OSes. These are the latest figures from
http://www.w3schools.com/browsers/browsers_os.asp

I often look here when OS and Browser usage is discussed.
And I agree, continued support for W2000 seems like an exaggerated benevolence.
Based on numbers only, support for Linux would seem in bigger demand.

2008 WinXP W2000 Win98 Vista W2003 Linux Mac
April 73.3% 3.3% 0.5% 8.8% 1.9% 3.7% 4.6%
March 72.7% 3.7% 0.6% 8.5% 1.9% 3.9% 4.4%
February 72.4% 4.0% 0.8% 7.8% 1.8% 3.8% 4.3%
January 73.6% 4.0% 0.8% 7.3% 1.9% 3.6% 4.4%

I also notice at the same source that Firefox is close to 40% of
browser usage, but we still see MSIE-centric web applications.

Cheers, Mats.E

2008/6/1 Chandler Coleman <chan...@yahoo.com>:

[Lars I. Nielsen (GisPro)]

Hi Mats,

This is a professional community last I heard. The figures you mention
are mostly related to private usage, imho. Can you point to figures
based on purely professional usage ?

And although I'm an avid and passionate Firefox user myself, I
acknowledge that the majority of professional browser use is still IE
based, because most companies run Windows. So IE-centric web solutions
are - unfortunately - something we must accept as part of the real world.

The point here is, that any software developer worth his trade uses
whatever his employer/customers use, or else he won't be able to target
their needs, and will be out of employment/business soon.

I assume that this is also the case in the MapInfo community ?

Best regards / Med venlig hilsen
Lars I. Nielsen
GIS & DB Integrator
GisPro

Mats Elfström skrev:

[Mats Elfström]

Hi group, and Lars.

I am sorry if I stepped on your professional toe.
I did not try to distinguish professional OS usage from any other
kind. As a matter of fact, I would not know how to define that.
Somebody asked for OS usage on viable systems. Such systems are likely
to be used on the Internet, and will leave their mark in the
statistics. I agree that the figures are mostly made up by home and
private users (as that is by far the majority) but do not think that
the percentages would differ greatly if one were able to sort out
different kinds of users.
If one were able to break down the numbers to MapInfo Professional
users only, the picture could of course change significantly, and I
guess that Pitney Bowes are using such statistics for their business
decisions.

The question about web solutions is another piece of pancake altogether.
The Internet is built on standards, and no web developer can know
which browser will call on his application.
Making IE centric solutions will shut out a very large number of
potential users, and that is not 'targeting their needs' by any means.
That is my view, perhaps shared by some, but I also know that many do
not agree.
I am reluctant to discuss the last point further on this forum. At
least, make a new thread if you feel it is important to do so.

Regards, Mats.E

2008/6/2 Lars I. Nielsen (GisPro) <L...@gispro.dk>:

[Warren Vick]

Hello Chandler,

 

I have one client who is a major telecoms company who is still on Windows 2000 and is only now in the process of moving to XP. Ditto for my bank. I would guess tens of thousands of systems between them and I’m sure the users consider them “viable”. While it may seem to most that Windows 2000 is very old, the wheels in the corporate world sometimes turn very slowly. I fully sympathise with organisation who have a huge installed PC based and, given the effort required to update, have to carefully consider the benefits of doing so. OS choices aside, I agree that .NET 2.0 was the right level for Pro integration.

 

Regards,
Warren Vick
Europa Technologies Ltd.
http://www.europa-tech.com

[Tim Rideout]

We still have about 3 PCs (25%) that are on Windows 2000, the rest are all XP, including the new ones bought this month. I am certainly not going to dabble with Vista until I am sure it all works without problems (and runs all our older software!).

 

I have wasted days if not weeks on upgrades etc that really bring little benefit.

 

Did you know that Illustrator CS runs faster on my 10 year old Windows 98 laptop with 64 MB of RAM than it does on XP on an AMD Athlon 64 with 1 GB and fairly new? At least until it runs out of memory, but start up is about twice as fast and you can work very nicely on small artwork.

 

Just for thought!

 

Tim

 

Dr Tim Rideout

Director

 

See the new Dynamic World wall map at http://xyzmaps.com/acatalog/World_Maps.html

Visit XYZ at the Cape Town Book Fair, 14-17 June 2008

 The XYZ Digital Map Company

Unit 9-11 Hardengreen Bus.Pk.
Dalhousie Road,
Dalkeith,
EH22 3NX

Tele: +44 (0) 131 454 0426
Mobile: +44 (0) 7766 825937
Fax: +44 131 454 0443
Email: tim.r...@xyzmaps.com
Web: www.xyzmaps.com

---

From: mapi...@googlegroups.com [mailto:mapi...@googlegroups.com] On Behalf Of Chandler Coleman
Sent: 01 June 2008 17:00
To: mapi...@googlegroups.com
Subject: [MI-L] Re: What's new in MapInfo 9.5