开云体育

On 9/8/2020 9:06 PM, Toby wrote:

On 2020-09-08 8:05 p.m., Harvey White wrote:

Hmmm

Well, I write a lot of my own programs, and databases for the programs
(assuming they use them) are easy enough when the programs use that
data, have a similar data structure, or I had enough time invested in
the program to make it worthwhile translating structures.

You have to design that backwards compatibility *into* the programs, if
it makes sense.

It's possible to design a program so different (not that you should)
that importing data from the last one makes no sense.

Having said that, there's a Kanji dictionary program that, if I ever
resurrect it, would be able to read the old data files.

Then again, I wrote that, and the effort would be worth the effort.

Yes, the key word is EFFORT, I didn't say it was impossible but it all
takes WORK to make happen, to migrate data, to make things futureproof,
to continue porting them forward year by year. And you're just b*ggered
if you're using proprietary stuff that didn't.

--T

Harvey



On 9/8/2020 5:18 PM, Toby wrote:

On 2020-09-08 3:32 p.m., Paul Amaranth wrote:

Gotta agree with Dave here.? Software can last forever.

Not without extraordinary effort.

Didn't really want to add to an off topic thread, but just preserving
_data_ is a hard problem. Keeping programs buildable and runnable is
even harder.

--T

On 9/8/2020 5:18 PM, Toby wrote:

On 2020-09-08 3:32 p.m., Paul Amaranth wrote:

Gotta agree with Dave here. Software can last forever.

Not without extraordinary effort.

Didn't really want to add to an off topic thread, but just preserving
_data_ is a hard problem. Keeping programs buildable and runnable is
even harder.

--T

On September 8, 2020 5:30:13 PM "Sandra Carroll" <smgvbest@...> wrote:

I'm going to drop most of this convo but this statement I can not leave alone.
" The reason they don't want z/OS source code released is so that the cracker kiddies don't see just how much it is Swiss cheese with regard to security."
I have no idea where you get these false facts from but they are indeed false.

And FWIW, I am a z/OS Security engineer and I was a Linux RHCE and Windows MCSE in the past.

Sandra Carroll

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Dave McGuire
Sent: Tuesday, September 8, 2020 1:23 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Looking for recommendations on 3D modeling software

On 9/8/20 12:55 PM, Sandra Carroll wrote:

Disagree,
Im talking about code out on github or sourceforge that are setup as you describe.
Projects die. I cannot quote the source exactly but where was an article that showed that over 70% if not higher Opensource project die.
I actually think it was like 90% but can not be sure, I know it was high.

Unless a project is widely adopted it will eventually die no mater how good it is.

If you're talking about "projects", sure, they die all the time. But the subject here is SOFTWARE. And making that assertion about software isn't just questionable, it's dead wrong, and I prove it every day.

I do this literally every day at the museum, and sometimes outside the museum. Hell, just last night I got a piece of HP software up and running that hasn't been supported since the 80s and I'm pretty sure nobody else has run since not long after that. Why? The source code was published.

The "project" is long since dead, but who cares? The software runs just fine.

This is something we do, very literally, every day at the museum, and I've done it over and over in other environments when the software wasn't quite so old.

As I inferred I'm a mainframe person. MVS was in todays terms open source. Infact you can get MVS 3.9 still as its under that license.
But at that time there was no github. No way to really share code. After taking so many calls asking to fix code companies wrote IBM decided to go OCO (object code only)
So they could support their customers needs consistently.

Fast forward to today. No one wants z/OS (formerly MVS) to go opensource. Most of its' customers are in financial. Security is extremely tight
Governments use it. None want z/OS to go opensource. And there's nothing wrong with that. Its want we the customers want and isnt' that the point.

The reason they don't want z/OS source code released is so that the cracker kiddies don't see just how much it is Swiss cheese with regard to security. That's a documented fact; z/OS binary distributions have been floating around illicitly for years and people have performed penetration testing on it. Information on this is everywhere. The best thing that could ever happen to z/OS would be to release the source code, so a few million young and communicative eyeballs, (rather than a few dozen old and isolated ones, could audit that code.

But of course that'd be a disaster, because of the security breaches that would occur first, after the flaws are discovered but before PTFs are rolled out.

And people have been sharing code for decades...wow, Sandra, github did not invent source code sharing! Heck, even in the somewhat stunted and isolated IBM mainframe world there was (and still is!) the SHARE group and the CBT tapes! Open source software has been the way of computing since there WAS computing. Suits rolling in after the fact and commercializing everything was, and is, still an outlier for many of us. There is, and always has been, far more open source software in the world than closed source.

IBM gave away their OSs at first because they didn't see any intrinsic value in software; they wrote it to sell hardware, and for no other reason. They closed it up when they realized they could make people pay for it, and thus create a new (and huge) profit center. This is a documented and well-known fact.

I'm sorry to be so disagreeable, but you're incorrect on so many levels here that I just can't sugar-coat it. It's nothing personal and of course I mean no disrespect whatsoever; I just have to correct your rather dramatic misconceptions.

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Dave McGuire
Sent: Tuesday, September 8, 2020 1:23 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] OT: Looking for recommendations on 3D modeling software

On 9/8/20 12:55 PM, Sandra Carroll wrote:

Disagree,
Im talking about code out on github or sourceforge that are setup as you describe.
Projects die. I cannot quote the source exactly but where was an article that showed that over 70% if not higher Opensource project die.
I actually think it was like 90% but can not be sure, I know it was high.

Unless a project is widely adopted it will eventually die no mater how good it is.

If you're talking about "projects", sure, they die all the time. But the subject here is SOFTWARE. And making that assertion about software isn't just questionable, it's dead wrong, and I prove it every day.

I do this literally every day at the museum, and sometimes outside the museum. Hell, just last night I got a piece of HP software up and running that hasn't been supported since the 80s and I'm pretty sure nobody else has run since not long after that. Why? The source code was published.

The "project" is long since dead, but who cares? The software runs just fine.

This is something we do, very literally, every day at the museum, and I've done it over and over in other environments when the software wasn't quite so old.

As I inferred I'm a mainframe person. MVS was in todays terms open source. Infact you can get MVS 3.9 still as its under that license.
But at that time there was no github. No way to really share code. After taking so many calls asking to fix code companies wrote IBM decided to go OCO (object code only)
So they could support their customers needs consistently.

Fast forward to today. No one wants z/OS (formerly MVS) to go opensource. Most of its' customers are in financial. Security is extremely tight
Governments use it. None want z/OS to go opensource. And there's nothing wrong with that. Its want we the customers want and isnt' that the point.

The reason they don't want z/OS source code released is so that the cracker kiddies don't see just how much it is Swiss cheese with regard to security. That's a documented fact; z/OS binary distributions have been floating around illicitly for years and people have performed penetration testing on it. Information on this is everywhere. The best thing that could ever happen to z/OS would be to release the source code, so a few million young and communicative eyeballs, (rather than a few dozen old and isolated ones, could audit that code.

But of course that'd be a disaster, because of the security breaches that would occur first, after the flaws are discovered but before PTFs are rolled out.

And people have been sharing code for decades...wow, Sandra, github did not invent source code sharing! Heck, even in the somewhat stunted and isolated IBM mainframe world there was (and still is!) the SHARE group and the CBT tapes! Open source software has been the way of computing since there WAS computing. Suits rolling in after the fact and commercializing everything was, and is, still an outlier for many of us. There is, and always has been, far more open source software in the world than closed source.

IBM gave away their OSs at first because they didn't see any intrinsic value in software; they wrote it to sell hardware, and for no other reason. They closed it up when they realized they could make people pay for it, and thus create a new (and huge) profit center. This is a documented and well-known fact.

I'm sorry to be so disagreeable, but you're incorrect on so many levels here that I just can't sugar-coat it. It's nothing personal and of course I mean no disrespect whatsoever; I just have to correct your rather dramatic misconceptions.

-Dave

--
Dave McGuire, AK4HZ
New Kensington, PA

On Tue, Sep 08, 2020 at 01:22:37PM -0400, Dave McGuire wrote:

On 9/8/20 12:55 PM, Sandra Carroll wrote:

Disagree,
Im talking about code out on github or sourceforge that are setup as you describe.
Projects die. I cannot quote the source exactly but where was an article that showed that over 70% if not higher Opensource project die.
I actually think it was like 90% but can not be sure, I know it was high.

Unless a project is widely adopted it will eventually die no mater how good it is.

If you're talking about "projects", sure, they die all the time. But the
subject here is SOFTWARE. And making that assertion about software isn't
just questionable, it's dead wrong, and I prove it every day.

On 9/8/2020 10:01 AM, Sandra Carroll wrote:

I’m surprised no one has mention fusion360. Full CAD/CAM with Eagle now as well.
3D parametric, direct manufacture to 3D printers. I send to both simplify3D and chitubox all the time.

Easy to use. Still no cost to makers.

On open source
Some comments. I’ve run into countless projects that die or are abandoned in the opensource community. Yeh some of the big stay around but far greater go nowhere.

Opensource is not this great pancea of perfection. If your not a developer yourself you can only hope they’ll accept your proposal for a change. Personal experience is they don’t always do this. In fact in my experience they rarely do. It does not fit the authors vision of how it should work.

I love Perl. But the community refused for the longest time to accept ebcdic(mainframe) into the their world. And only did by making it ascii internally which has its own problems for those of us in the mainframe.

I’m not against opensource, I use it all the time. Like anything it has it’s upside and downside.

Sandra

Sent from my iPhone 7 Plus

On Sep 8, 2020, at 9:29 AM, Chuck Harris <cfharris@...> wrote:

?Bear in mind that one of the features of open source projects
is they really never go away. Commercial projects come and vanish,
and change their licensing terms with great regularity.

OpenSCAD is used by dozens of other projects, and is a very solid
package. The source is available on github, and if you don't like
the direction the authors are going, you can take it and make it
your way. Sure, if you aren't talented that way, you will have to
hire it done, but that is what you are already doing with the outright
commercial packages... without any hope of customization.

If OpenSCAD is too hard to learn (it isn't), you can use one of the
many packages, like FreeCAD, that have incorporated it into their
graphic interfaces.

We greybeards remember the PC so well, that we wrote GNU and linux
for it, and emancipated ourselves from Microsoft.

Educator licenses are available to universities to allow them to
train their kiddies on the package at no cost to the university.

The university professor gives a 1 year copy to each student, and
they use it for their course work. At the end of the year, it is
inactivated.

The companies do that because it is free advertising, and a free
beta testing of their product. They also know that once trained on
a product, students will ask for it where they work.

That is the sole and only reason Apple and Microsoft have such nice
licensing terms for K-12 schools.

Also, to use a student license, usually you are required to show
registration at a university. The rules on student licenses are
highly restrictive relative to commercial intention. A product
designed on a student license cannot ever be commercialized without
transferring it to a full on commercial seat... at typically $50K a
seat... and once there, it can never be brought back to a student
seat.

-Chuck Harris

Tom Gardner wrote:

Overall I doubt that members of this group will want to make anything really complex,
but they might like it to be available to other members in a decade or so. So have a
quick look and choose any tool that feels comfortable without a large learning curve.

After a///very//quick look/, it appears that OnShape is....

Free for "educators", with a 1 year time limit on the licence - whatever that means.

Online only, running in a browser like OpenJSCAD. The standard questions with any
"cloud service" are whether:

* it will be there in 5 years time; see Microsoft PlaysForSure(TM), and giggle
* the licence conditions can be changed, e.g. is the company/product changes ownership

Greybeards will remember the sighs of relief when PCs became available, because it
meant that users' data was not "held hostage" inside silos owned by other companies.

On 08/09/20 02:54, Kuba Ober wrote:
OnShape - as long as it’s not for commercial purposes - all your projects are
publicly viewable then. A joy to use. All it needs is an OK internet connection and
a supported web browser.

FreeCAD is another option.

7 sep. 2020 kl. 10:17 fm skrev victor.silva via groups.io
<daejon1@...>:

?This is off topic but the end result will be for HP products so please bear with me.

I am looking for recommendation on 3D modeling software (preferably freeware) and
a company that I can then send the 3D model to
make a battery holder clamshells. I propose to make half clamshell pieces that
would then make a complete
battery holder by using 2 pieces that would fit together.

Thank you,
Victor