Impulse Tracker source code now available on GitHub | Hacker News Reader

Impulse Tracker source code now available on GitHub

github.com

200 pointsbane43 comments

The Wikipedia entry on text-based user interfaces uses IT as an example of a text-mode program featuring a pixel-accurate mouse pointer. The "citation-needed" had always been annoying me, since I knew it was true, but I couldn't prove it. Now, we can finally link to IT_MOUSE.ASM and be done with it.

Bisqwit has a video about the technique: https://youtu.be/7nlNQcKsj74

But is Impulse Tracker a true textmode application? While not full of flashy graphics and icons, it uses lots of widgets and layout features that put it squarely in conventional GUI territory. And the repo contains the VESA code for rendering the equalizer UI.

Apparently this was hidden behind the Alt+F12 key combo? I might have to fire up a copy as I don't remember ever seeing this bit. Edit: it doesn't seem to be enabled in 2.14. Oh well, got to hear "Blue Flame" again!

The rest of the application runs in 80x50 text mode, and it included a character editor so you could customize the box drawing characters if you were so inclined.

It was a common technique in the DOS era to simulate GUIs with text. This was not just done with the characters in codepage 437 either; using redefinable character sets in the EGA and VGA, graphical elements such as buttons, input fields, etc. could be rendered in character cells. By redefining characters where the mouse pointer was supposed to be, a pixel-accurate mouse pointer could be created. Not just trackers used this technique, but e.g., later versions of Norton Utilities did as well.

That seems like an analytical conclusion based on a primary source, and thus a violation of No Original Research policy if sourced directly to the primary source on which the conclusion is based, and not a published secondary source that actually provides the analysis of the primary source.

[citation needed]

Just cite that HN comment above you then;)

One of Wikipedia's sourcing paradoxes is that the literal horse's mouth is the worst source as to be unusable.

I mean this as a genuine comment and not a political non-sequitor but we'll see how it lands. If you think that's a bad idea, can you imagine what Donald J. Trump's wiki page would look like if you allowed 'primary' sources? Or, really, any political figure. Though I think he makes a very good implicit case against allowing that.

"Original research"/primary sources should not matter for facts, i.e. those laid out as actual code.

(It didn't land well.)

I agree. It doesn't seem to make sense for some facts.

I was way out of my wheelhouse on Wikipedia one day and I saw some popular K-Pop group had announced their new album release date on their Instagram feed. I updated their Wikipedia entry with a citation to the group's own announcement and was very swiftly smashed by some over-arching editor who told me what a fuck-up my life was for even thinking this might be a suitable source for the information.

If the fact is in the source cited, primary sources are okay as direct sources. If it is an analytical interpretation of the source (such as most fact claims about the source or its contents, rather than fact claims made directly in the source), then that interpretation needs sourced from somewhere.

Somewhere that isn't the person updating the article, that is. The trick is to make a random website that does the same, and then it's totally fine!

No, primary sources are fine if they say the thing you are sourcing them for (high quality secondary sources are usually preferred, but primary sources aren't “the worst” or “unusuable”); they aren't okay for something that requires additional analysis/interpretation. (That's true of any source, but it’s more of an issue with the use of primary sources.)

Note that this source code was originally released back in 2014 on bitbucket as a mercurial repository. Looks like he’s now moved it to GitHub. Here’s the original announcement http://roartindon.blogspot.com/2014/10/20-years-of-impulse-t...

As far as I can tell, the LICENSE.TXT (https://github.com/jthlim/impulse-tracker/blob/main/LICENSE....) is the 3-clause BSD license, meaning proper FOSS and not just "source available" which is becoming more popular.

Bit more about the history of IT (and more) in the blogpost celebrating the "20th anniversary", written by the software's author: http://roartindon.blogspot.com/2014/02/20-years-of-impulse-t...

> Copyright <YEAR> <COPYRIGHT HOLDER>

https://opensource.org/license/BSD-3-clause

10 years ago when publishing the code to bitbucket, I grabbed what I thought was a copy of the BSD-3-clause and replaced my name and year

Seems like it wasn’t accurate. I updated the license on the github repository a few hours ago so it is exactly the BSD-3-clause template now

Nice! I was a Fast Tracker 2 guy back in the day. More recently I've played with OpenMPT and MilkyTracker (both of which I packaged for PortableApps.com) with my old MODs. I love seeing older stuff get properly open sourced.

Impulse Tracker: https://en.wikipedia.org/wiki/Impulse_Tracker

Fast Tracker 2: https://en.wikipedia.org/wiki/FastTracker_2

Any thoughts on Hively Tracker?

I'll take a look at it.

Have you checked out 8bitbubsy's extremely accurate modern clone of FT2 (and one of Protracker as well)?

https://16-bits.org/ft2.php

I just did! That's awesome. I may need to package these, too.

They're already available in the Debian repos and probably elsewhere as well.

Oh I meant package for PortableApps.com.

Thanks for milky on portable apps! Use it all the time

You're welcome :)

Fast Tracker 2 in javascript: https://www.a1k0n.net/2015/11/09/javascript-ft2-player.html

IT was probably one of the most influential pieces of software for me as a teenager. I can’t put into words how important it was to me as a way to escape a crappy life and just get lost in making music and sharing tracks on my local BBS, chatting to other people like me who I’d go on to meet and become life long friends with.

It will always hold a special place in my heart and I owe Jeffrey so much.

Me too! I grew up very isolated in rural America and IT was a literal gateway into a global artistic movement that caused me to connect to people across the planet, develop a skill, and set me on a course into technology. I am forever in Jeffrey's debt.

It’s hard to understate the impact IT had on the tracker scene. Thanks, Jeffrey!

Wow, check out https://github.com/jthlim/JavelinPattern ... guy's been busy.

That's impressive!

Great software, great sound. Here is a tune I recommend.

https://youtu.be/zuQHHlbzX_c

Sweet jesus, I knew IT had been written in assembly (and moreover, it looks like 8086-level assembly, with EMS support for memory expansion), but little did I knew it had support for networking! Granted, IPX, but still!

From the Wikipedia page:

> extra features, such as support for stereo WAV output and a personalized version of the driver for co-editing songs over IPX networks, were provided for a fee.

Also looks like tfa should have a (2014). Ed: Ah, wikipedia mention bitbucket and links to an unofficial GitHub mirror, different from TFA.

On a related note, I'm a little shocked to discover that MED/OctaMED apparently has new/current commercial support?!

https://forum.amiga.org/index.php?topic=76507.0

An interesting coincidence was that custom IPX usernames was a paid feature, and payment was processed by a company named "Kagi" (not the search engine).

https://github.com/jthlim/impulse-tracker/blob/fb47b325454ab...

Kagi the e-commerce site apparently went defunct in 2016:

https://web.archive.org/web/20160915172120/http://kagi.com/

I thought you meant Kali until I checked the link.

https://en.m.wikipedia.org/wiki/Kali_(software)

I hope hope hope the SchismTracker folks take a hard look through this!

What would be the reason for that, other than general interest of how things were done? Any effects or playback differences that still need to be ironed out?

Reply on news.ycombinator.com