Development

Stuff

Steady on chaps; we all know windows is the best thing since computers were invented; so why all the advocacy?

Shame it doesn't run on the pi, though.

Let's stick on-topic and not ignite the 'doze fanbois.

It is clear to me that you only want a dumbed down system on which you click with the mouse and everything magically installs.

In this context I don't understand why you keep comparing installation of prepared Windows packages with compiling free Linux software from source. Those are different things and they always will be.

Why with Windows and not with Linux?

This is not a problem of Linux, it is a problem of the hardware manufacturer who chooses to write Windows drivers and not Linux drivers, and of you buying hardware without checking for support first.

I use NVidia video cards, the manufacturer supports Linux, writes and releases a driver, and listens when people complain about problems.

You know what, they even jump through the hoops that the fanatics of free software hold up for them to make life difficult for hardware companies who want to support Linux without giving away their soul.

When it is your feel that Windows supports so much more hardware, then what makes you end up here? I presume Windows supports the Raspberry PI, so why not run Windows on it?

This is the same for Windows and Linux. And for Linux there exist even more free development environments.

When you think it is too difficult to install packages on Linux, then simply use Windows. It makes no sense to claim that things are too difficult when they simply are different from what you have used for a while. I work with Linux and Windows, and Windows may be easy for those who want to install some standard program or hardware, it certainly is not as easy to administer as most people think it is, and the results of this show up on the internet all the time. (trojans, botnets etc)

On 09/06/2013 09:51, Rob wrote: []

If you only consider stand-alone programs your figure may be right. I suspect that quite a few spreadsheet users and others do code macros, though, which could be considered programming. As one aim of the RPi is to encourage programming, likely Linux is a reasonable choice.

Although a lot of the "programming" I see offered is not what I would call "real" programming, just simple graphics calls and perhaps event-driven responses. Better than nothing.

try getting X windows to work on Windoze....

The point is what is supplied works. What is not supplied at least CAN be made to work without spending MONEY.

If you have a GUI on the linux there are many good text editors for programming. I used pluma and Geany for example.

Console wise, I use vi, because by and large I don't do much at console level. And years of programming on UNIX systems taught me that the only thing I could guarantee to find would be vi. I hate it, it is an total abortinon, but I CAN DRIVE IT. Ive written hundreds of thousands of lines of code in vi. I still hate it. All teh more because I dont use it that often and have forgottien most of its commands.

If I am running a headless server, the very first things I do is share its drives via NFS or samba and mount them on a machine with a proper gui, so I can in fact edit the source at least on a machine I am familiar with.

Then use a console session from the machine via telnet or ssh to actually do the compilations. If you make a Makefile, that generally amounts to typing 'make' anyway..

All yu need thereafter is to remember the error messages spewing from te console and link them to the source file that just failed to compile..

Huh?

Huh?

Maybe so, maybe not. There's a discussion on this point running on comp.risks. The emerging consensus seems to be that creating spreadsheets should be classified as a programming task, but that variety of incorrect and untested spreadsheets (and real-world errors arising from them) show that too many spreadsheet authors don't have the skills and training that's needed.

Quite. The RPi was designed to be used as a development platform rather than as a small, cheap device for running COTS packages.

exactly. I have four networked together here over two geographical locations.

-- Ineptocracy

(in-ep-toc?-ra-cy) ? a system of government where the least capable to lead are elected by the least capable of producing, and where the members of society least likely to sustain themselves or succeed, are rewarded with goods and services paid for by the confiscated wealth of a diminishing number of producers.

The advantage of termcap is that I can add any new, previously unknown capability to termcap I want by writing it into a terminal definition in / etc/termcap and using it in a program. No other changes needed.

I can't do that with terminfo because it has a predefined capability list. If the capability I need isn't defined in tic, tough. I can't use it because the position and size of capability values in a compiled terminfo definition were set when tic and libterminfo were compiled.

I'd remind you, too, that there would appear to be no publicly available definition of the compiled module format and that different implementations use different compiled formats.

He means "the main problem with linux, is that it wont install and run windows apps"

This is true and it is a drawback.

But if its a show stopper, don't run linux. Or use it on an X86 platform and run virtual box and windows inside it.

Linux had a fully functional network stack back when Windows users were still playing with "Trumpet Winsock"...

But that is only because you are familiar with other programs. When I am on a different platform and I want to modify some piece of code, the first thing is to see if I can find a port of vi.

When I want to add another IF to a block of code, and need to increase the indent on that block, I know how to do it in vi in two or a few keystrokes, and in programs like notepad or wordpad this cannot be done at all or requires repetitive insertion of spaces.

That is the kind of thing a programmer needs. Not a search and replace for fixed words but a regular-expression search and replace with capture of substrings and use in the replacement.

I can understand why features like that are not built into programs written for mom and dad, but as I programmer I use them all the time and I need them in the system default editor.

I don't forget the commands but that probably is because I learned them while I was in my twenties. And I like the fact that I don't need to grab the mouse all the time while I am working. When I browse the internet I use the mouse, when I type text I use the keyboard. I don't like programs that mix the use of the mouse and the keyboard.

On 09/06/2013 13:20, The Natural Philosopher wrote: []

PuTTY and Xming do that.

"In 1982, the Internet protocol suite (TCP/IP) was standardized, and consequently, the concept of a world-wide network of interconnected TCP/IP networks, called the Internet, was introduced. Access to the ARPANET was expanded in 1981 when the National Science Foundation (NSF) developed the Computer Science Network (CSNET) and again in 1986 when NSFNET provided access to supercomputer sites in the United States from research and education organizations. Commercial Internet service providers (ISPs) began to emerge in the late 1980s and early 1990s. The ARPANET was decommissioned in 1990. The Internet was commercialized in 1995 when NSFNET was decommissioned, removing the last restrictions on the use of the Internet to carry commercial traffic."

Let's see when windows 1.0 came out...

"Windows is announced in 1983, but it takes a while to develop. Skeptics call it ?vaporware.?"

They still do..

It had no networking beyond Novell Netware until Windows for Workgroups in 1993.

Linux predates it being launched in 1991. Complete with a stack.

So Windows was about 11 years behind the Unix curve on networking. It didn't have its own IP stack until windows 95.

And two years behind Linux.

I do this by adding what I need to the terminfo description and compiling it. See

where the extensions are colored differently (kLFT3 for instance).

so do most programs using termcap or terminfo.

I don't recall a previous reminder, but updated the manpage a while ago.

Here's a link (aside from the extensions, term.5 is quite old):

In practice, there aren't many different implementations, and ncurses can be compiled to support the others (to make it interoperate better on a given host). Again, done a long time ago, the most recent is noted here:

In anything that's been maintained over the past 15 years (s-lang and NetBSD), it's using the bit-layout that ncurses uses - but lacking the extensions done in 2000.

The easiest releases of Linux didn?t include any networking support - not even AF_UNIX IPC. The first release to support TCP seems to have been 0.98, which contains timestamps up to late September 1992.

Should be ?earliest?.

I don't consider Windows an OS until the release of Windows 95.

X isn't an OS either.

TCP/IP networking for Windows 3.x was introduced by a third party developer "Trumpet", also in 1993 I think. The original Windows for Workgroups networking did not use TCP/IP.

I started using Linux in 1992 and I was immediately able to connect it to the VAX/VMS system at work using TCP/IP and X. I could use it as a graphical desktop system. I had color while the existing users of the system had only black-and-white terminals.

We had SCO XENIX and Unix systems as well but the networking and X were expensive options for them, we used serial terminals instead.