ejolson
Posts: 2867
Joined: Tue Mar 18, 2014 11:47 am

Re: Is ARM doomed?

Sat Mar 30, 2019 10:05 pm

Heater wrote:
Sat Mar 30, 2019 7:53 pm
If somebody could fix the the appalling state of programming tools for us experienced adults that would be great. We muddle through of course but it must be billions of professional programmer man hours wasted on that stuff.
Those who remember how simple things used to be are too busy making up for lost man hours. Those with the time to fix things are likely responsible for the way the tools currently work. Once the next generation takes charge, new tools will be created that work exactly like Scratch, except for the orange cat.

And you ask, is ARM doomed?

User avatar
Gavinmc42
Posts: 2883
Joined: Wed Aug 28, 2013 3:31 am

Re: Is ARM doomed?

Sat Mar 30, 2019 11:11 pm

Haha. We are all doomed.

I do like the idea of Scratch, it is just offensive to graphics trained me that the pieces don't fit together.
In some places those jigsaw pieces look like they should not be there
And it is way too slow to use.
The many versions of "blocky" seem to repeat the same ugliness.

And why do we have to do it online or only on Windows or Mac.
Browser only coding?
Next version on tablets and Smartphone?

Next gen of coders, "I have no touch screen so I cannot program"?
Already I have seen students going "I could not find the answer on Google so could not do the assignment".
Or the other "I don't want to do that project, it has programming in it".

Yep we are doomed if this is the next gen.
Hurry up with AI computers that program themselves?
appalling state of programming tools
Not talking about bloatware like Eclipse by any chance?
When I spend more time learning and fighting the tool than programming I tend to move onto something easier.
When my first choice of tool is Geany, I'm either not doing enough coding or don't feel the need for something more complex.

Been coding since the 1980's, you would think after 4 decades things would be simpler, not so complex that it takes years to learn the tools.
The tools should be simple and stay out of the way.
And you really only need debuggers when using C, higher level language compilers don't let as many errors though, flame on ;)
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

jahboater
Posts: 4183
Joined: Wed Feb 04, 2015 6:38 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 9:22 am

timrowledge wrote:
Sat Mar 30, 2019 7:31 pm
Scratch is an environment for introducing people with no prior experience to the concept of programming a computer.
Scratch is brilliant for that. I used it in exactly that way to demonstrate to someone what programming was about. Brilliant.
timrowledge wrote:
Sat Mar 30, 2019 7:31 pm
Instead of dumping newcomers into a nightmare of stupid text editors and command line with ridiculous incantations
Two different problems. Scratch is not a replacement for these software tools.

People who complain about current software tools should write something better or be quiet. Given how clever the original authors were, I suspect that even the design of such better tools would be a significant challenge.

I don't actually like using "vi" myself but I recognize that the design is extremely clever and a skilled user can do more work with less key strokes than with any other editor. So I don't complain about it.

Small steps.
Emacs is extremely powerful and extensible. But to extend it (which many people do) you have to learn Lisp.
Perhaps produce an editor as powerful as emacs that can be extended with a popular modern language (Python for example).

Actually, editors are easy to write. Its the design that is hard.
Last edited by jahboater on Sun Mar 31, 2019 11:08 am, edited 1 time in total.

jahboater
Posts: 4183
Joined: Wed Feb 04, 2015 6:38 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 9:42 am

Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
Been coding since the 1980's, you would think after 4 decades things would be simpler, not so complex that it takes years to learn the tools.
Software naturally "grows" with time. More and more features get added. So its bound to get more complex.
The trick is to learn just enough to get by. I only use a tiny percentage of the power of current tools.
I recently purchased Richard Stallman's book on GDB:
https://www.amazon.co.uk/Debugging-GDB- ... oks&sr=1-1
It is a very large book indeed, I had no idea GDB was so powerful.
Still only use a few commands though.

User avatar
Gavinmc42
Posts: 2883
Joined: Wed Aug 28, 2013 3:31 am

Re: Is ARM doomed?

Sun Mar 31, 2019 10:41 am

Software naturally "grows" with time. More and more features get added. So its bound to get more complex.
The trick is to learn just enough to get by. I only use a tiny percentage of the power of current tools.
I was only using a tiny <2% of even a B+'s power, seemed a waste, but it was easy to get on the network and the app was mostly shell script.
But it was using Linux and Busybox, the ideal target for DDoS ;)
More complex translates to more vulnerable?

Brooks said a good programmer can write 1000 lines of bug free code per year.
If I do 10 projects a year then each one would have to be 100 lines of code ;)
And I'm not even a good programmer :oops:

Don't remember who Tannenbaum was quoting or the exact quote but it was about simplifying and taking just the right amount of stuff off but no more.
Why are Retro computers popular with us older coders? Because they were simpler?
The old days of one person writing an OS or a Game, have they gone.
Can we get that back with Pi's?

Is ARM doomed because they keep adding stuff that no one person can figure it all out?
Is that why some of us oldies are looking at RISC-V?
Can we go to simpler cpus, just lots of them?

Remember the old cars we could fix ourselves because we knew how they worked.
Look under a hood these days, not even room to get your hand and spanner in there.

So is ARM doomed because it takes 100 people to write code for it and no one know how it all works?
500 man years on the VC4, been told lots of luck writing my own image sensor pipeline :lol:
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 10:59 am

Gavinmc42,
But it was using Linux and Busybox, the ideal target for DDoS
No. How on Earth could that be?
More complex translates to more vulnerable?
Quite likely. See Windows...
Brooks said a good programmer can write 1000 lines of bug free code per year.
Brooks did some surveys and is more famous for determining that a typical programmer produces in the order of 10 lines of code per day. After accounting for design, meetings, documentation, meetings, testing, meetings, etc over the life of a project. As many as 2000 per year!

I'm not sure if it was Brooks but others have documented that there can be in the order of 10 bugs in a thousand lines. So a programmer create 200 bugs per year.

ejolson
Posts: 2867
Joined: Tue Mar 18, 2014 11:47 am

Re: Is ARM doomed?

Sun Mar 31, 2019 12:10 pm

Heater wrote:
Sun Mar 31, 2019 10:59 am
So a programmer create 200 bugs per year.
The bugs are not important as long as they run fast so nobody notices.

Last night my friend was telling me about a software design method where one team is tasked to add features with merit based on lines of code written while a second team is tasked to simplify things with merit based on lines of code removed.

When answering questions about doom, maybe there should be a part of government specifically tasked with simplifying the legislation written by the others. As for ARM, it is notable that instruction sets come and go, rather than just piling on top of each other.

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 12:39 pm

ejolson,
Last night my friend was telling me about a software design method where one team is tasked to add features with merit based on lines of code written while a second team is tasked to simplify things with merit based on lines of code removed.
Tell me more. I'm curious to know what the result of such a method might be.

If a team member of the first team met up with a team member of the second team over beers they could come up with a scheme to earn themselves tons of merit!

User avatar
davidcoton
Posts: 3699
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: Is ARM doomed?

Sun Mar 31, 2019 4:17 pm

Heater wrote:
Sun Mar 31, 2019 12:39 pm
ejolson,
Last night my friend was telling me about a software design method where one team is tasked to add features with merit based on lines of code written while a second team is tasked to simplify things with merit based on lines of code removed.
Tell me more. I'm curious to know what the result of such a method might be.

If a team member of the first team met up with a team member of the second team over beers they could come up with a scheme to earn themselves tons of merit!
Define "line of code".
It is possible to adapt one's programming style (certainly in C, C++, Python) to produce more or less lines to implement the same idea. Often without changing the algorithm, just writing it differently.
Or perhaps, copy in some function instead of using a library call.
LOC is a very poor metric of programmer productivity or merit. It is an unreliable guide to complexity. I thought the LOC obsession was over when I was still a professional software engineer, 15-20 years ago. The sooner any remaining ideas about linking LOC with program or programmer metrics is killed off, the better. It is such practices that "doom" coding projects.
Signature retired

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 4:59 pm

davidcoton,
Define "line of code".....
Yes. I think we all know that lines of code is a pretty sloppy metric for programmer productivity. For all the reasons you mentioned above and more.

I'm sure we can all tell anecdotes of times when we spent a month on something and ended up with less code than we started with that performed better and was easier to understand and maintain into the future. Our bosses did not fire us for negative productivity.

As you say, productivity, as in rate of functionality implemented, will vary greatly with different styles, languages, and so on.

I think the problem is that in those past decades you mentioned nothing better has taken it's place. Not anything that works across language and application domains etc. It's the best we have.

And, Brook's estimate of 10 lines per day per programmer is still true. As is the observed bug rate per line. It seems to be independent of language or programmer skill.

As it happens, out of curiosity, I have sometimes downloaded largish projects from github for which the development time and approximate number of developers is known. Sure enough on counting the lines, Brooks was right! Try it for yourself.

I conclude that given typical developers working on typical projects, who are not gaming the system, LOC still is useful as measure of the size of a problem you have.

jahboater
Posts: 4183
Joined: Wed Feb 04, 2015 6:38 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 6:30 pm

Heater wrote:
Sun Mar 31, 2019 4:59 pm
I think the problem is that in those past decades you mentioned nothing better has taken it's place. Not anything that works across language and application domains etc. It's the best we have.
Well said.

David Coton,
Can you suggest any better metric?

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Sun Mar 31, 2019 7:21 pm

jahboater,
People who complain about current software tools should write something better or be quiet.
I can appreciate that. As such can I downgrade what I said from "complaint" to "observation"?

My observation is that many of today's development tools are obscenely huge, complex and slow. Demanding a lot of effort to find your way around them. In short, time wasting and tedious. It's not entirely clear to me that all of that is necessary.

Further, it's not just the editors and IDEs. This issue stems from the very languages, compilers, run times and build systems. It's systemic. It's not something I or any regular developer can fix to satisfy himself, far too big a task.

For the sake of brevity I'll restrict my complaint, err observation, to two cases that have been bothering me this weekend and wasted many hours of my life:

1) Altera's, now Intel's Quartus dev tools for FPGA:

If you want to toy with FPGA you need Quartus or similar HDL dev tools from the FPGA vendor. Oh my God, they are gigabytes of download and install, they take forever to become familiar with, the actual editors are terrible and the edit, build test cycle is mind bendingly slow. Sure they are very feature packed. But really.

As it happens we now know this need not be the case as Clifford Wolf and friends have produced FPGA tool chains for Lattice FPGA's that are orders of magnitude smaller and faster whilst also being very simple to use.

2) Scala, Java, IntelliJ, SBT...

If you have toyed with FPGA a bit you soon find that the hardware description languages, Verilog and VHDL, are a pain. So I thought I try something a bit more high level, SpinalHDL. Now Spinal is a Scala library. That means getting familiar with Scala. But Scala is not a real language, it compiles to Java bytecode and depends on the Java libraries. That means needing all that stuff. The whole concoction is so complex that it's almost impossible to write code without an IDE to guide you, that pulls in InteliJ and SBT as the build system.

Before you know it you are up to your eyeballs in complexity and having a hard time making the pieces work together. And of course compilation is tediously slow (That is not actually compilation, just generating Verilog from VHDL from Spinal/Scala). And noticing that a simple job requires 3GB of RAM!

Luckily again we now know this need not be so. Someone has a plan to fix this. Per Vognsen has made great progress in creating a high level HDL in Python. Not done yet.

3) C++

Oops, sorry, said I stick to two complaints/observations.

User avatar
davidcoton
Posts: 3699
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK

Re: Is ARM doomed?

Sun Mar 31, 2019 11:47 pm

jahboater wrote:
Sun Mar 31, 2019 6:30 pm
Heater wrote:
Sun Mar 31, 2019 4:59 pm
I think the problem is that in those past decades you mentioned nothing better has taken it's place. Not anything that works across language and application domains etc. It's the best we have.
Well said.

David Coton,
Can you suggest any better metric?
If I could, I would be in the consultancy business making a fortune from my discovery.

But the lack of a "better" (read: working) metric should not prevent us fromthrowing out a broken one -- though I concede that there are certain limited things the LOC does indicate on a large scale, and where the coding standards and styles are comparable.

I found that moving to an object-oriented development model allieviated the problem, and overall gave much better productivity. This is because the programming phase (where LOC are SO tempting as a metric) is reduced to a much smaller time segment, when properly sequenced with requirements capture, design, analysis, (coding) and test. An alternative metric based on requirements identified/designed/analysed/coded/tested has some merit, but neds some weighting for the size of the requirement. The major problem was that senior managers still wanted early completion estimates (before the project scope was understood), and would not take on trust that they would get later but more accurate estimates in place of the previous (non-OO) estimates which were known to be unreliable.

YMMV.
Signature retired

ejolson
Posts: 2867
Joined: Tue Mar 18, 2014 11:47 am

Re: Is ARM doomed?

Mon Apr 01, 2019 1:55 am

davidcoton wrote:
Sun Mar 31, 2019 11:47 pm
jahboater wrote:
Sun Mar 31, 2019 6:30 pm
Heater wrote:
Sun Mar 31, 2019 4:59 pm
I think the problem is that in those past decades you mentioned nothing better has taken it's place. Not anything that works across language and application domains etc. It's the best we have.
Well said.

David Coton,
Can you suggest any better metric?
If I could, I would be in the consultancy business making a fortune from my discovery.
For modern languages, better metrics are easy to come by:

. For LISP count parenthesis.

. For C count semicolons.

. For Go count missing semicolons.

. For Python count white space.

. For Scratch number of mouse clicks.

. For Java size of the byte code.

. For BASIC sum of the line numbers counted negative.

. For C++ same as C except a bonus 500 for every use of friend and operator.

. For COBOL, FORTRAN and Assembly obviously use the number of lines.

Milliways
Posts: 389
Joined: Fri Apr 25, 2014 12:18 am
Location: Sydney, Australia

Re: Is ARM doomed?

Mon Apr 01, 2019 2:34 am

I don't understand the fuss over RISC - this is old hat.

The IBM 360 used microcode in 1964 to implement its instruction set on a RISC-like processor.

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 3:18 am

Milliways,
I don't understand the fuss over RISC - this is old hat.
I don't think there is a fuss over RISC. Pretty much everyone has come to agree that that is the way to design fast, efficient processors. Everyone uses RISC, even if it's hidden away internally like Intel's machines. Nobody has designed and built a non-RISC processor for three decades or more.

The old "fuss" about RISC vs CISC was I believe more of a marketing war between CPU vendors than an actual debate of design merits. As you say, that debate, if we can call it that, is old hat.

I opened this thread with RISC V as a topic. I will suggest that the "RISC" aspect of RISC V is not the important point. The main point of RISC V is to have an open and free to use instruction set architecture. To garner support for a common standard that is not proprietary. Thus giving freedom to academia, start ups and others to get things done more reliably, economically and efficiently.

To that end it makes sense that the design has a simple basis and makes use all the the lessons learned in ISA design over decades. That rather dictates a RISC machine.
The IBM 360 used microcode in 1964 to implement its instruction set on a RISC-like processor.
And is therefore, in combination, not a RISC design. Arguably a prime example of CISC.

ejolson
Posts: 2867
Joined: Tue Mar 18, 2014 11:47 am

Re: Is ARM doomed?

Mon Apr 01, 2019 3:35 am

Milliways wrote:
Mon Apr 01, 2019 2:34 am
I don't understand the fuss over RISC - this is old hat.

The IBM 360 used microcode in 1964 to implement its instruction set on a RISC-like processor.
The story of IBM is arguably filled with more doom than any other computer manufacturer--mostly because the company has survived for so long, but also because of the mindset that set in after the antitrust case long ago.

Even though IBM now has a very small market share, they have not been able to shake their former fear of becoming too successful. Now the question is how much smaller can IBM shrink before Samsung or another buys them for the patent pool. Maybe the RedHat acquisition will energise the company and turn things around, especially if ARM is as doomed as the title of this thread suggests.

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 4:30 am

ejolson,

IBM is of course where the whole RISC idea grew from. They are still at it with the Power Architecture I believe.

What a great idea, never mind ARM and Power, IBM is a founder member of the RISC V Foundation. Let's have an IBM RISC V and RedHat.

With all that openness all the world would love them again and there might finally be an IBM product I might like to own :)

jahboater
Posts: 4183
Joined: Wed Feb 04, 2015 6:38 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 8:49 am

ejolson wrote:
Mon Apr 01, 2019 1:55 am
For modern languages, better metrics are easy to come by:

. For C count semicolons.
I prefer LOC because:

1) it includes much more of the language (for example all the conditional logic "if( expr )")
2) correct comments take time to write and are essential
3) blank lines may improve the readability of the code
4) trivial to count with "wc"!
Last edited by jahboater on Mon Apr 01, 2019 8:59 am, edited 1 time in total.

User avatar
Gavinmc42
Posts: 2883
Joined: Wed Aug 28, 2013 3:31 am

Re: Is ARM doomed?

Mon Apr 01, 2019 8:57 am

Has someone done a metric.

Bugs per lines of code verses language?
Bugs per lines of code verses CPU architecture?

Or should that be lines of code per bug? :lol:
No weird stuff, has to be readable and understandable code.

Lines of Comments per lines of Code?
OS verses known bugs?
Came a across an OS that claims no known bugs, unfortunately my brain has a memory leak and has lost this wonder.
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 9:44 am

Gavinmc42,
Has someone done a metric.

Bugs per lines of code verses language?
In deed they have. People have been trying to collect such statistics for decades. The observation seems to be that the bug rate is about the same no matter what language is used. See for example the work of Steve McConnell.
Bugs per lines of code verses CPU architecture?
Interesting idea.

Given that high level languages abstract away the instruction sets it would not be very meaningful there. But what about assembly language programs? I guess there are not huge amounts of assembler around so it would be difficult to get statistics.

But what about the actual CPU architectures? Now a days logic, inc. CPU's, are designed using hardware description languages: VHDL, Verilog, Bluespec, Chisle etc. That is to say a CPU design looks like software, we can count LOC. Clearly processor that takes more LOC to define it will have more bugs.

Back in the day the new Intel 286 processor had hundreds of bugs. We had a huge document from Intel, under NDA, describing them all. At the same time the new ARM chip from Acorn probably had none due to it's simplicity.
Lines of Comments per lines of Code?
Comments are of course to be avoided in ones code.

If your code is so badly organized with such bad variable, function, class, method names that it needs lots of comments to describe what it does and how it works, then fix the code.

Comments are invariably incorrect. That confuses future maintainers causing them to introduce bugs.

We see that comments are as critical to your products correctness as the actual code. If not don't put them in. As such they should be included in LOC. We know that more LOC means more bugs. Ergo decrease LOC by minimizing comments.
Came a across an OS that claims no known bugs...
So did I. It would not let anyone log in :)

Of course, none of my code has any known bugs either.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 22061
Joined: Sat Jul 30, 2011 7:41 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 10:12 am

Heater wrote:
Sat Mar 30, 2019 7:53 pm
Despite my jibes I do appreciate the idea behind Scratch, Tim.

Having spent most of today fighting Scala, InteliJ, SBT and Quartus I particularly feel the point about stupid IDEs, text editors, build systems and so on.

Still, I recall young teenagers in the early 1989's building robots with their Sinclair Spectrums, so I find it amazing we need all the complexity something like Scratch to achieve the same ends, perhaps less, today.

If somebody could fix the the appalling state of programming tools for us experienced adults that would be great. We muddle through of course but it must be billions of professional programmer man hours wasted on that stuff.
So stop the jibes. Really.



As for optimisations, we have spent quite some times optimisation stuff on the Pi. So its a worthwhile skill, despite the fact that compilers are awesome, but micro optimisations aare only necessary for a tiny minority of cases. For example, one of our engineers spent some time optimising a software HEVC decoder. He was looking to save only two or three cyles in a loop which was the difference between 720p60 working or not. Its hardcore NEON code. The algorithm is fixed, so its purely down to coding and trying to find cycles. Rare, but it still happens.

Most humans only need to select the correct algorithm, then let the compiler do its job.

With regard to metrics, I've recently come (almost) to the end of a job to fix up a load of preexisting code to work better with multiple framebuffers. From git show of all the changes, grepping on "+ " which is the diff for inserted lines I get about 1200 lines changes, when I grep for ";" I get about 1500. Neither are particularly accurate, but given its taken about 12 days of work, that's about 100-120 lines of code per day. About ten times the metric suggested above, which I am quite familiar with. So does the 10 lines per day metric still apply to modified code, rather than code generated from scratch?
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

User avatar
Gavinmc42
Posts: 2883
Joined: Wed Aug 28, 2013 3:31 am

Re: Is ARM doomed?

Mon Apr 01, 2019 11:15 am

Most humans only need to select the correct algorithm, then let the compiler do its job.
Most people don't know what an algorithm is ;)
Why can't we get the computer to pick it own algorithm?
Yes I know some have done it, just not my Pi's, yet.

Hope that engineer got a medal, maybe a badge he can sew on his shirt, free mug?
RPT got one of those clocks on the wall, Number of days since last bug was killed?
Plot bugs killed on daily chart? Is Tuesday the best day?

I think that no known bugs was in a nano or micro kernel OS, which has the drivers in user space so that does not count ;)
AROS? I think I remember they said it only had 4000 lines of code in the Kernel?

Edit google know- Minix3 ;)

If the Silicon is simple it would be easy to spot errata?
It has to be 64 bit so that rules out the 6502.
Why 64 bit?
So it can hold everything in memory, all the worlds information can have an address.
Well that's according to Tannenbaum or was it some one else he quoted?
Hurry up with those memory implants, mine keeps leaking.

Bug day Clock would make a nice Pi project.
How about a Doomsday clock for the various cpus?
Z80's are still ticking along.
When year was peak 6502?
What year will the ARM's run out?
I'm dancing on Rainbows.
Raspberries are not Apples or Oranges

Heater
Posts: 12137
Joined: Tue Jul 17, 2012 3:02 pm

Re: Is ARM doomed?

Mon Apr 01, 2019 8:33 pm

I thought for a moment this is perhaps this is an April 1st joke:

A Memory Segmentation Extension Proposal for RISCV: https://www.youtube.com/watch?v=nkk5HA4LtoY

timrowledge
Posts: 1257
Joined: Mon Oct 29, 2012 8:12 pm
Location: Vancouver Island
Contact: Website

Re: Is ARM doomed?

Wed Apr 03, 2019 11:51 pm

Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
I do like the idea of Scratch, it is just offensive to graphics trained me that the pieces don't fit together.
Hmm, as a graduate of the Royal College of Art I'd agree - except that in 'real' Scratch as installed in Raspbian as "Scratch" rather than "Scratch 2", they *do* fit together
Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
And it is way too slow to use.
I'd agree, mostly, there. These sort of graphical manipulation languages are never the fastest way to express some computation for anyone that can already handle text languages. BUT (and it's a big wobbly butt) they provide a way for people that are *not* yet that capable to do interesting things. I've never yet seen one that can decently handle really complex situations either; any of the old "show a network of all the connections" type rapidly turned into bad underground maps.
Just like Pascal, it is not intended to be a serious real-world programming system, but an intersting way to teach.

Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
And why do we have to do it online or only on Windows or Mac.
Browser only coding?
Next version on tablets and Smartphone?
I hate the 'browser only' approach of Scratch 2 & 3. It isn't a nice idea even on a big fast machine that can cope with the insanity of 'modern' web browsers - but I do see the problem it was trying to alleviate whereby in many school districts the teachers & students are not able to install anything without official help. Of coursse, this was irrelevant on a Pi, where Scratch was already installed.
Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
appalling state of programming tools
Not talking about bloatware like Eclipse by any chance?
Eclipse is an abomination in the eyes of Nuggan.
Gavinmc42 wrote:
Sat Mar 30, 2019 11:11 pm
Been coding since the 1980's, you would think after 4 decades things would be simpler, not so complex that it takes years to learn the tools.
The tools should be simple and stay out of the way.
And you really only need debuggers when using C, higher level language compilers don't let as many errors though, flame on ;)
That's too daft to be worth flaming...
Making Smalltalk on ARM since 1986; making your Scratch better since 2012

Return to “Off topic discussion”