Computerphile I hope you are aware you are creating a highly important archive of computer science knowledge from someone who 'was there' - You are doing what the BBC and other broadcasting networks which cover science are failing to do. Thank you so much.
My favorite computerphile speaker!! You can hear the passion and excitment in his voice, which makes it so fun to listen to! I truly appreciate all your work, the whole team. I'm doing my Phd in philosophy with computer science as an AOC, and this channel has helped me soooo much!!
Man, I think this man deserves his own UA-cam Chanel! Something like 'Storys from the past of computer technology' It's always nice to listen to him, would love to have more content with him...
I love these explanations from Professor Brailsford, he has such a soothing tone in his voice! ;-) Anyhow, since you're digging into computer history, I'd really like to know where the usage of pointy brackets for doing something came from, back in the days of the FIDO net (the FIDO net may in itself be worth a whole computerphile video!). They were later, in the late 90s on the www replaced by stars. So when I got "online-socialized", you'd write something like or , while later on people wrote (asterisk)ROTFLOL(asterisk) and (asterisk)duck&run(asterisk), sometimes leaving out the end asterisk, and nowadays things like are just written without any indicating characters for the special type of "chain of letters" that is not to be read as a normal word, but as something the writer is doing, or an acronym of that. How did this change come to pass, and how was it "invented" in the first place? It goes along with the first smileys ": - )" (w/o the spaces), but somehow I have never found a good explanation of how it all came to be that way, and the changes that were made to this "informal notation" over time. I'd really like to hear someone (preferrably Prof. Brailsford, if he knows anything about it) talk about this - maybe in conjuncture with the times of mailboxes and the FIDO net and so on....this is an important part of computer history, *especially* ("old style" asterisk usage - notation for bold text here - fortunately adopted by the youtube/G+ comment system) since it was a decentralized way of spreading information digitally, which is something we almost don't have any more today, because everyone uses a platform or at least a web host company for their data to be spread...
More than anything right now, I'd really love to see a Computerphile video on the power of Regular Expressions. Prof Brailsford would be great at explaining this to people who don't know about it.
Sgml was a precursor to html. Used primarily to structure document formats so they could be machine independent. In 1992 I wrote my masters thesis in SGML. And many defence contracts require technical documentation in SGML so that they can be consistently read on any computer. I ran a chunk of these contracts up to 2010.
Im a big fan of computerphile. I'd like to make a suggestion for a future episode on the topic of procedural generation, specifically in video games. I really like how you guys get experts, to speak about subjects that they are expert about. I think this topic would greatly benefit from this. Theres a million videos on this topic, but most of them are just made by gamers, not by computer scientists that can talk about the theory. Keep up the good work computerphile!
Even though I am going to live in a way more technological suffocated world them him, I am so jealous that he was one of the many men who essentially help created the amazing world of computers we have today
Charles also talks about the reduction of 'keyboarding' as a reason why the end-tag could be omitted. IOW, the parser, the software that reads and processes the SGML markup can infer the end of a given element, the element at the top of the open element stack, based on the next element encountered. Or not for an end-of-input condition.
A story I heard was that at the first demo of a Fortran compiler, an error was generated 'Missing close bracket on line 50'. Someone in the audience said why didn't Fortran insert missing bracket? This is why HTML etc allows missing closures
It's so obvious that learning becomes profoundly easier when the first generation of knowledge holders educate on a topic. As the generation of knowledge holders withers down through time, a lot of important details are lost in teaching, thereby creating a set of less informed students and practitioners.
+GeirGunnarss That was cruel, it was like getting a "to be continued" at the best part of a movie! I'm looking forward for the second part. I really enjoy Professor Brailsford's way of teaching.
That American military initiative, CALS, was "Computer-aided Acquisition and Logistic Support" then later "Continuous Acquisition and Life-cycle Support." No wonder you couldn't come up with the name off the top of your head. Making up acronyms is a fun game in the civil service. I attended a few conferences.
It would be easy to codify the omitted closing tag by simply enforcing an assumption. If a closing tag is encountered while (an)other tag(s) are open at a deeper scope, those open tags are also considered closed. So for example John says, Bring back my bike. Since the Q tag is at a deeper scope than the P tag, when the P tag is closed, the Q tag is also considered closed. Then you wouldn't need any specific rules for each document, it'd just be baked into the general rules of the protocol.
+3ZYBRE +Gregory Sherman I'm pretty sure that XML came out of SGML and is a more generic multi-purpose and formalised extensible superset of the original concept, removing the more ambiguous features. I go that from his description, but then I'm extremely familiar with XML.
+m4tt3m50n Personally I prefer Json to XML these days. Easier to read and write, once you learn the rules and syntax. A lot easier to use to serialize data as well.
+Richard Smith For many intends and purposes json is indeed awesome. Mostly for its simplicity, easy parsing and grammar rules, readability, and small footprint. But XML has it's place too, I'd argue that xml tends to stay cleaner when more complexity is introduced. The consistent syntax for parameters, subcontent, namespacing, etc can really help with managing large data. Additionally, json schemas are fairly new and adoption is still quite low. Xml schemas on the other hand are mature, and there's decent tooling around them. Even though they're a royal pain to write :P.
Seems like once you were writing data, you could strictly speaking make an implementation of SGML where you just abruptly end the file, then have the program as it is running and working with your file, close the tags FILO style and save disk space that way. In the case of the memo, you could just skip closing the q, p, body and memo tags. If it is important to know when the file or string ends, could maybe just close the memo tag, but keep the q, p and body open.
The US Department of Defense (DOD) had a long-term project to reduce the cost of supporting and constructing equipment used by the military. Using the acronym CALS, standing originally for Computer-aided Acquisition and Logistic Support and then for Continuous Acquisition and Life-cycle Support, the DOD developed a family of standards for digital information of various types. According to A Brief History of the Development of SGML,
Before SGML there was GML. Generalized Markup Language. The G was Charles Goldfarb. M Moshe. And L Laurie (sp?). Goldfarb wrote the book in the video. Where are M and L?
This is making me wonder if you could use a markup language to tell a program how a binary file it's going to read is arranged, allowing you to support new formats that use existing features without needing to recompile the program to make it support them.
So to recap sgml is a rule set for defining your own markup languages and what is legal and not legal within the confines of said created markup language.
+Locut0s Yep! That's just about it. The only problem is that there was an unwritten rule, among those in the know, that the "ML" suffix should be reserved for meta-languages like SGML and XML and should NOT be used for specific *applications* of these meta-syntaxes. Hence the deprecation of Tim B-L's naming of his Web page tagset as "HTML".
Computerphile I hope you are aware you are creating a highly important archive of computer science knowledge from someone who 'was there' - You are doing what the BBC and other broadcasting networks which cover science are failing to do. Thank you so much.
Same here :) lol
??
Damn I love listening to the man talk.
+Franklin Cerpico I completely agree. His excitement and emphasis for what he talks about drags me in.
He is a teacher that could literally teach anybody properly.
+1
+kubi spice Just think that some furtunate people have been his students... lucky guys huh?
the David Attenborough of computerphile
My favorite computerphile speaker!! You can hear the passion and excitment in his voice, which makes it so fun to listen to! I truly appreciate all your work, the whole team. I'm doing my Phd in philosophy with computer science as an AOC, and this channel has helped me soooo much!!
Watching this for the first time tonight and I wanna join whatever class I assume he teaches!
Damn you cliffhanger! I could've sat for hours and hours listening to Brailsford, his passion and quiet enthusiasm get me hooked.
Never actually heard of SGML before this video.
+LFalch me neither
+ByteBitTV 3 people commenting dont know what it is
+Its Glizda that makes 4 of us learning something new today :)
+LFalch I've heard of it, but only vaguely as some predecessor of XML.
+thoyo ++ ...Make that 5
This guy is the spitting image of my late grandfather, and he talks about cool computer stuff, it's like a dream come true!
Each video with Professor Brailsford is such a pleasure to watch! Thank you very much.
false.
Love this guys voice, I want him to read me bed time stories
I'm scared.
ok?
Man, I think this man deserves his own UA-cam Chanel!
Something like 'Storys from the past of computer technology'
It's always nice to listen to him, would love to have more content with him...
I love these explanations from Professor Brailsford, he has such a soothing tone in his voice! ;-)
Anyhow, since you're digging into computer history, I'd really like to know where the usage of pointy brackets for doing something came from, back in the days of the FIDO net (the FIDO net may in itself be worth a whole computerphile video!). They were later, in the late 90s on the www replaced by stars. So when I got "online-socialized", you'd write something like or , while later on people wrote (asterisk)ROTFLOL(asterisk) and (asterisk)duck&run(asterisk), sometimes leaving out the end asterisk, and nowadays things like are just written without any indicating characters for the special type of "chain of letters" that is not to be read as a normal word, but as something the writer is doing, or an acronym of that. How did this change come to pass, and how was it "invented" in the first place? It goes along with the first smileys ": - )" (w/o the spaces), but somehow I have never found a good explanation of how it all came to be that way, and the changes that were made to this "informal notation" over time. I'd really like to hear someone (preferrably Prof. Brailsford, if he knows anything about it) talk about this - maybe in conjuncture with the times of mailboxes and the FIDO net and so on....this is an important part of computer history, *especially* ("old style" asterisk usage - notation for bold text here - fortunately adopted by the youtube/G+ comment system) since it was a decentralized way of spreading information digitally, which is something we almost don't have any more today, because everyone uses a platform or at least a web host company for their data to be spread...
More than anything right now, I'd really love to see a Computerphile video on the power of Regular Expressions. Prof Brailsford would be great at explaining this to people who don't know about it.
The professor has such a soothing voice, I really enjoy the videos with him.
Sgml was a precursor to html. Used primarily to structure document formats so they could be machine independent. In 1992 I wrote my masters thesis in SGML. And many defence contracts require technical documentation in SGML so that they can be consistently read on any computer. I ran a chunk of these contracts up to 2010.
Came here to find out more about SGML in a video, thanks very much for presenting the information in a very understandable way :).
You shall not omit the end of the video!
+Simon Vetter No worries, I've got the fix right here.
Nooo this isn't good at all! What have we done?!
??
This guy is such a joy to speak with!
Im a big fan of computerphile. I'd like to make a suggestion for a future episode on the topic of procedural generation, specifically in video games. I really like how you guys get experts, to speak about subjects that they are expert about. I think this topic would greatly benefit from this. Theres a million videos on this topic, but most of them are just made by gamers, not by computer scientists that can talk about the theory. Keep up the good work computerphile!
What an awesome type of person he is!! I'd never mind going to college if he was my professor!!
computer scientists will mourn the day this man passes on to the far interweb
such an amount of knowledge of science and history of computers
Feels like I’m getting a history of computer science lesson, and I love it.
Brilliant man! Great discussion!
Professor Brailsford videos are the best!!
Thank you for remembering all these important bits of history!!!
Even though I am going to live in a way more technological suffocated world them him, I am so jealous that he was one of the many men who essentially help created the amazing world of computers we have today
XML vs JSON episode please! One of my favorite discussions ...
Whaaat, you can't just stop in the middle of the story like that! This was super interesting!
Charles also talks about the reduction of 'keyboarding' as a reason why the end-tag could be omitted. IOW, the parser, the software that reads and processes the SGML markup can infer the end of a given element, the element at the top of the open element stack, based on the next element encountered. Or not for an end-of-input condition.
I'm in the middle of pee, now I'm gonna do another pee without finishing of the first pee. My standards compliant toilet will still accept it.
ok?
wow I did not expect that the markup languages that we take for granted nowadays had such history behind them.
Great instruction! That made way more sense than my textbook, thank you.
A story I heard was that at the first demo of a Fortran compiler, an error was generated 'Missing close bracket on line 50'. Someone in the audience said why didn't Fortran insert missing bracket?
This is why HTML etc allows missing closures
Chrome and other browsers add closing tags
Really enjoyed this one. Looking forward to the HTML one. Now let's talk XSLT. :)
It's so obvious that learning becomes profoundly easier when the first generation of knowledge holders educate on a topic. As the generation of knowledge holders withers down through time, a lot of important details are lost in teaching, thereby creating a set of less informed students and practitioners.
Ugh. That cliffhanger!
??
Great video, fun to learn about this history as a web dev.
his talking is so easy to listen to ! i'd like to have more lectures of him.
It's cool how all computerphile videos start with and end with
I'm really loving this. OAO
Can't wait for the next one!
Moar Professor Brailsford! I love him! Amazing passion and style!
This had as much RDFa as XML. Excellent video introduction to SGML. Thank you.
0:41 the dramatic pause after "lawyer" lmao
Nooooo! Don't leave us hanging like that! that was fascinating. More.
+GeirGunnarss Biggest tease ever, right!
Nova Fawks Well, i could mention some girls from high school but he is damn close to it, yeah. Hehe. :)
+GeirGunnarss My exact same reaction... I even shouted it out loud.
+GeirGunnarss That was cruel, it was like getting a "to be continued" at the best part of a movie! I'm looking forward for the second part.
I really enjoy Professor Brailsford's way of teaching.
??
Love your histories Professor !!
I could listen to this for hours
Screw you end-tag, I need my precious bits. 😂 [t.stamp 04:56]
Great explanation, thank you. DeltaXML have some really useful XML compare and merge tools with API for building into applications.
And thus, XHTML was born!
That American military initiative, CALS, was "Computer-aided Acquisition and Logistic Support" then later "Continuous Acquisition and Life-cycle Support." No wonder you couldn't come up with the name off the top of your head. Making up acronyms is a fun game in the civil service. I attended a few conferences.
This was so fun to watch 😆
This man is the David Attenborough of Computer Science.
Is this man one of the inventors of HTML?!
Great vid! Shame I missed it when it came out! I was subbed too!
haha I was just about to scroll down to the comments and look for "you missed the and then he stopped me
I dig this guy......such interesting historical experiences
I love this channel.
Amazing person to listen! Thanks!
It would be easy to codify the omitted closing tag by simply enforcing an assumption. If a closing tag is encountered while (an)other tag(s) are open at a deeper scope, those open tags are also considered closed. So for example John says, Bring back my bike. Since the Q tag is at a deeper scope than the P tag, when the P tag is closed, the Q tag is also considered closed. Then you wouldn't need any specific rules for each document, it'd just be baked into the general rules of the protocol.
Why do you need speech marks in addition to quote tags? Surely they could be redundant too to save more disc space.
I love to listening this man ❤️ thank-you
Oh! A cliff hanger! I am waiting for part 2.
Am I wrong to insist that a video actually cover what the title says it covers? It should be titled SGML to HTML - How we got here from there"
+Gregory Sherman Yeah, I didn't hear much about XML...
+3ZYBRE +Gregory Sherman I'm pretty sure that XML came out of SGML and is a more generic multi-purpose and formalised extensible superset of the original concept, removing the more ambiguous features. I go that from his description, but then I'm extremely familiar with XML.
You are very right. This video was interesting, but not really finished, which is annoying.
+m4tt3m50n Personally I prefer Json to XML these days. Easier to read and write, once you learn the rules and syntax. A lot easier to use to serialize data as well.
+Richard Smith For many intends and purposes json is indeed awesome. Mostly for its simplicity, easy parsing and grammar rules, readability, and small footprint.
But XML has it's place too, I'd argue that xml tends to stay cleaner when more complexity is introduced. The consistent syntax for parameters, subcontent, namespacing, etc can really help with managing large data.
Additionally, json schemas are fairly new and adoption is still quite low. Xml schemas on the other hand are mature, and there's decent tooling around them. Even though they're a royal pain to write :P.
Love his enthusiasm
Glad you gave IBM credit (DCF)
that was very interesting. looking forward to the follow up
I had a google search about SGML and after 10 minutes, now I’m here
hello
Seems like once you were writing data, you could strictly speaking make an implementation of SGML where you just abruptly end the file, then have the program as it is running and working with your file, close the tags FILO style and save disk space that way. In the case of the memo, you could just skip closing the q, p, body and memo tags. If it is important to know when the file or string ends, could maybe just close the memo tag, but keep the q, p and body open.
the title of this episode should be "SGML, HTML, XML. What's the difference. Are they all the same? Let's find out!"
Love the explanation :)
CALS -> Continuous Acquisition and Life cycle Support.
Engaging presentation. Wish you were my lecturer!
this is gold, shohld be shown to people starting woth web development imho
The US Department of Defense (DOD) had a long-term project to reduce the cost of supporting and constructing equipment used by the military. Using the acronym CALS, standing originally for Computer-aided Acquisition and Logistic Support and then for Continuous Acquisition and Life-cycle Support, the DOD developed a family of standards for digital information of various types. According to A Brief History of the Development of SGML,
and what about xml? did I miss something or did he not talk about it?
ARGH! CLIFF HANGERS! X{
??
Before SGML there was GML. Generalized Markup Language. The G was Charles Goldfarb. M Moshe. And L Laurie (sp?). Goldfarb wrote the book in the video. Where are M and L?
Can you make a video about how languages like C or Javascript came around?
This is making me wonder if you could use a markup language to tell a program how a binary file it's going to read is arranged, allowing you to support new formats that use existing features without needing to recompile the program to make it support them.
This guy is great!
Nah, I just closed the joke proper so it won't break the page. :)
+Joakim Rannikko haha, the geekiest banter ever, i love it!
??
XML and friends (XPath/XSLT) is just an enterprise version of Lisp (more specifically Scheme/Lisp-1)
I was almost say ''You missed out the tag , hahhah
Great videos thanks!!
He would be a great narrator for fantasy RPGs (both video games and pen & paper)
Only Thunderbird does right. Outlook is terrible in representing quoted text.
Love this guy
No, please not another story on another day! It's a bit soul crushing to hear that...
And this is why, in Emacs, html-mode is derived from sgml-mode.
brilliant thank you
what is the spec /speck?
:( no clear answer to the question yet?
3:02
"Body part" ...
So to recap sgml is a rule set for defining your own markup languages and what is legal and not legal within the confines of said created markup language.
+Locut0s
Yep! That's just about it. The only problem is that there was an unwritten rule,
among those in the know, that the "ML" suffix should be reserved for meta-languages like SGML and XML and should NOT be used for specific *applications* of these meta-syntaxes. Hence the deprecation of Tim B-L's naming of his Web page tagset as "HTML".
Can you make a video about Petya and how it could be cracked?
so is HTML an application of SGML
+Antony D'Andrea
Yes! Absolutely! But very confusing that the both of them end in "ML" when they are very different things ....
Does anyone else think in this man's voice?
hope the next one discusses how verbose these are.
Why did you bother mentioning XML in this video?
They could have used s-expressions and saved a ton of time and space. Imagine how much power is consumed parsing closing tags around the world.