8 ms·
When I developed the first native C++ compiler around 1987, I thought I'd better check with AT&T's lawyers if I could: 1. sell a C++ compiler 2. call it C++
by WalterBright 4y ago
When I developed the first native C++ compiler around 1987, I thought I'd better check with AT&T's lawyers if I could:
1. sell a C++ compiler
2. call it C++
Their lawyer was very nice, and said sure. He also laughed and said he appreciated that I was the only one who bothered to ask.
- effingwewt 4y agoStill boggles my mind reading things like this- so cool to hear origin stories from the horse's mouth, so to speak. But if he hadn't been nice what would have happened? I'm sad to imagine the future we are headed towards where only new technologies are worked on, just to get the patents and sit on them, while everything pre-existing is deprecated and abandoned. Microsoft's embrace extend extinguish works eerily well. Do you think we are headed in a better direction now? Or is it simply different? I remember working with ATT in the late 90's when it was ATT broadband. They were still using green screen computer terminal windows for everything and were trying to get a GUI off the ground. Everything was alt and tab to switch windows etc, no mouse support! Into the mid 2000's! This was done by getting regular employees to try and cobble something together. The miraculous thing was that they pulled it off. Just took them a few years longer than it should have. ATT 'corporate' was known for being every bit the tv trope of a big business back then. I remember seeing early builds and laughing it was so bad. ATT in particular was very much stuck in its ways as a corporation and was not inclined to change even the most broken of things. When it finally rolled out it was such a mess. All the IT people could do was wince. Everyone but the people who mattered saw it coming. They hired some firm to re-do it all. That happened who knows how many times within just a few years after that. Went from refusing to update to doing it constantly and always breaking things. I must say it was pretty insane watching $12k long distance bills from calls to, for instance India, get re-rated to several dollars. Wondered how accurate the accounting could possibly be with so many inaccurate and fungible dollar amounts floating around. Edit to add- years later I was working as an HVAC service technician and had to do some work at cell phone towers, server locations (wasn't really server farms back then), and phone agent locations for all cell phone carriers in Albuquerque, NM. It felt surreal to see bow fast and far the companies had gotten in only a few short years. Companies (especially ATT) that couldn't figure out basic things about computers were at the cutting edge of like all the technology they used. ACs were top of the line Lieberts, they had all proprietary software on everything. The super remote cell phone towers had AC, power, storage, and communications redundancy. Their security had been beefed up to top tier. Contractors all needed top tier security clearance now. I don't know if it was the Kevin Mitnick generation of phone phreaks, hackers, and social engineering, or the world in general, or a change in CEO but it was kind of like watching the titanic become some super-advanced space-faring time-warping ship. That had kinda given me hope that- damn, maybe we can enter the new age jumping in with both feet. As a kid it felt like progress had been so slow!
- foobarian 4y agoWhat do you mean by "trying to get a GUI off the ground?" Was it that ATT corp IT was trying to deploy a windowing OS onto employees' computers? Or that they were developing a graphical OS a la Plan9 and failing?
- dekhn 4y agoOne of my favorite tech books is 'The Idea Factory' which covers various periods of innovation at (AT&T) Bell Labs, including the creation of the first real cell phone technology. When I first got Pacific Bell (now part of AT&T) DSL in California in the early 2000s, it was run by a seperate division of the company, "Emerging Products Division". i always assumed that was because the traditional side of PacBell just didn't get digital at all and the leadership kept them apart to avoid the innovator's dilemma.
- gumby 4y ago> When I first got Pacific Bell (now part of AT&T) DSL in California in the early 2000s, it was run by a seperate division of the company, "Emerging Products Division". i always assumed that was because the traditional side of PacBell just didn't get digital at all and the leadership kept them apart to avoid the innovator's dilemma. This technique had previously been used by IBM to the the PC out: they built a whole division from scratch in Boca Raton away from the IBM mother ship in NY
- ianai 4y agoI definitely wish more recent tech had gone the standardized way similar to how IBM standardized computer building. We’re into four decades of being able to build PCs from customized/off the shelf parts because IBM didn’t go the “make it impossible”/proprietary route.
- com2kid 4y agoDon't forget that Microsoft maintained momentum in this area. MS loved having open standards because it let MS pit OEMs against each other, causing hardware prices to drop while Windows license prices stayed the same.
- WalterBright 4y agoP.S. Bjarne Stroustrup and especially Andrew Koenig were super nice and supportive of my efforts in those days. The nascent C++ community was very welcoming.
- jacquesm 4y agoAnd yours was a huge improvement over 'cfront'! https://en.wikipedia.org/wiki/Cfront https://en.wikipedia.org/wiki/Cfront vs https://en.wikipedia.org/wiki/Digital_Mars https://en.wikipedia.org/wiki/Digital_Mars Did you get into trouble because of the Zorland joke?
- WalterBright 4y agoZorland was a mistake. Philippe was well within his rights to threaten a lawsuit over it.
- cgriswald 4y agoIn the early internet days, when I was young and learning HTML, the book I used to learn recommended asking permission to link to a website. I thought that was weird, but I was a rule follower then. So I, a kid, sent a message to the 'webmaster' of a local government site and dutifully asked permission to link to their home page. The said no. If I'm ever a super-villain, I'm using this as my origin story.
- WalterBright 4y agoOn the other hand, I've consistently found you can often get what you want just by asking for it. A lot of people seem unwilling to do this.
- megablast 4y agoThanks for all the examples.
- zarmin 4y agoOn the other other hand, ask for forgiveness not permission.
- WalterBright 4y agoMore than one person has come to me with a story that they'd developed a side project without notifying their employer, and now that it was done and they were distributing it were worried their employer was going to claim ownership. I didn't have anything helpful to say. I've done side projects when I was working for Big Corps, but in every case notified management beforehand and got a written ok. Never had any trouble with it. When I've accepted job offers, I'd also provide a list of projects that were mine and had them sign off on it as a condition of employment. Never encountered any resistance to that, either. But these poor people were sweating bullets imagining all the bad consequences of their employer finding out. Just ask, in advance. If you're a valuable employee, they'll say ok. Never heard of one saying no. And they'll appreciate that you asked instead of sneaking around. But be careful not to use company equipment.
- donarb 4y agoI recall that Microsoft's Visual C++ user license had a clause that forbid you from creating a C++ compiler with it.
- eschneider 4y agoI believe that was for your protection.
- bluedino 4y agoSounds like a challenge
- dylan604 4y agoHere, hold my beer! or Hide and watch!
- WalterBright 4y agoMore sales for Zortech C++!
- MaxBarraclough 4y agoNot a silly question at all, though. The situation isn't the same with Java, where expensive approval is needed before you can call your solution a JVM. The Kaffe VM, for instance, is careful to show a disclaimer: technically, Kaffe is not Java. [0][1] [0] http://www.kaffe.org/ http://www.kaffe.org/ [1] https://softwareengineering.stackexchange.com/a/433377/ https://softwareengineering.stackexchange.com/a/433377/
- jhgb 4y ago> I was the only one who bothered to ask. Was it necessary to ask? Unless C++ is patented and you live in a place where software patents are actually a thing, what could prevent you from writing your own implementation of something?
- WalterBright 4y agoI had no idea what IP rights AT&T held over it. There was no way I was going to invest blood, sweat, and tears in developing a C++ compiler only to get sued over it. Besides, it was polite to ask.
- benjamoon 4y agoI didn’t do anything as amazing as you did, but in about 1999 I made a website with a nav menu “inspired” by the then current Microsoft website nav (it was all vbscript dhtml and I basically copy and pasted it from the ms site). I emailed them to check it was ok to copy their nav (it wasn’t even a commercial project so I don’t know why I was so polite), but they replied and they were really nice. They said they encourage “engineers” to create new things rather than copy, but anything I needed from their site I could take and then a load of links to dhtml resources in case I wanted to build my own. I wonder if you did that now would you just get a response from their legal team?
- rkuykendall-com 4y agoOne of the first things I ever did with significant complexity in HTML was stealing the Apple.com Nav bar. When you're just starting out, that is a challenge all it's own!