[i336_]

I read through the recent thread on teaching kids to code (https://news.ycombinator.com/item?id=13499626) and was mildly perplexed at how seemingly successful Scratch has been.

It wasn't until the 2nd or 3rd time I'd used it that I actually figured out how to make sense of it and run something (for Scratch's definition of "run").

To be honest I've progressed extremely slowly with CompSci/programming over the past 18 years I've been using them (got my first computer around 7-8) - I started with QBasic, been shouting at PHP for way too long, I have a basic understanding of C I badly need to develop, and I'm moving toward playing with Lua next - and I hardly consider myself a dyed-in-the-algorithms academic type with a brain that's unable to understand Scratch. (In fact, I'd argue that the best programming teachers would be precisely those types of people, and if they were unable to understand Scratch that would be a major problem.)

Rather, I firmly believe Scatch's UI is a disaster, and horribly unintuitive to use. Other languages are beset with grammatical idiosyncrasies; with Scratch you have to learn the UI before you can learn the... few parts of the language that are actually there.

I'm concerned that systems like Scratch are so widely used; I fear that it's an even worse mind-scrambler than the bad sides of BASIC. Of course, like BASIC, there are good sides, and it teaches the basics without presenting a Mt. Everest-sized learning curve. Perhaps https://en.wikipedia.org/wiki/Dartmouth_BASIC was the Scratch of 1964, and I'm just griping about the dilutory effects of educationally-targeted software in this day and age and "modern" GUI design.

Scratch is also really slow/laggy on my old laptop (Thinkpad T43), I can't imagine how bad it is for schools with limited hardware.

[qwertyuiop924]

When I used Scratch 1.4, I don't recall the UI being too bad, and it ran pretty fast: Smalltalk is pretty good at that. But I'm generally pretty good at picking up these sorts of things, and my computer wasn't particularly slow.

If you want a Real Language presented the same way, Snap! (descended from BYOB) is essentially a Scheme in Scratch's clothing.

But the two real draws of Scratch were its hackability and its community. Back before Scratch 2.0 ruined everything, Scratch was written in Smalltalk, and using a widely-known hidden feature, you could examine the source code and make whatever changes you wanted with relative ease, resulting in a healthy community of mods and derivatives which explored new features and ideas, or those that the official team had dropped by the wayside (like Mesh, a fully-featured networking system).

Scratch's community was likewise excellent: I spent a lot of time lurking in the Scratch Advanced Topics forum - a sort of off-topic general programming section, where people far smarter than I discussed modifying Scratch, improving the website, and whatever programming projects they happened to be working on (usually web programming in PHP - it was the mid 2000s, after all).

But that's enough nostalgia for one day...

[i336_]

Ah, I encountered Scatch 2.0. I'll definitely check 1.4 out, it looks a lot more accessible and reasonable. I would have loved to have encountered something like this at 14 or 15.

I suspect the reason Scratch felt slow to me is that 2.0 is some kind of HTML5 and/or Flash mess now - you're right, Smalltalk is really fast. I spun up Squeak to check something on this T43 yesterday, and everything was really snappy. I have no reason to expect Scratch will be any slower.

Also, I wouldn't be surprised if a reasonable bit of the exploration everyone did was motivated by the fact that they were "hacking" the platform :P

There seems to be a sad lack of excellent online communities nowadays; I've long looked for sites to complement HN, but without success.

I just had a look at Snap! which is interesting. It definitely flatlines this laptop though, I had to try it on a faster machine. But I'm running the tree animation demo right now, and it looks awesome....

[qwertyuiop924]

>Also, I wouldn't be surprised if a reasonable bit of the exploration everyone did was motivated by the fact that they were "hacking" the platform :P

Nor would I. Even at the age of 8, before I was really able to understand the code, there was a thrill to it, in a cracking-open-the-toy kind of way.

And it helps that Smalltalk does exploration better than just about any other language/environment. You can just open up any Smalltalk app and extend/take it apart using the same tools the developers did to build it.

>There seems to be a sad lack of excellent online communities nowadays; I've long looked for sites to complement HN, but without success.

Lainchan (a sort-of cyberpunk/whatever chan) is quite popular with some of the people who are here on HN. It's a very different atmosphere, but it does emphasize actual good discussion. And it's got a containment board for politics, which always helps.

At the very least, their magazine (https://lainzine.neocities.org) is worth looking at, if not for the generally interesting articles, than for the outright strangeness of a lot of it.

[i336_]

Wow, nice! At 8 (1999) I was given a probably-6-or-7-year-old 286 running DOS 3.3 with nothing on it. That got swapped for an XT a couple years later, which I discovered Qbasic on and got tangled up in for way too many years :S. Smalltalk would have been awesome to discover at that age, moreso a toy with sekret doors and passages in it for me to discover :D

I was recommended Lainchain a couple months ago, actually, but nobody mentioned the magazine, which is really cool. I am not impressed that the ASCII art generation paper in Vol.1 has any associated source code!! The rest of the magazine content and design is really interesting too.