I'm surprised to see that this submission is from 2008; while I identify with the author's struggle (and subsequent delight) as he plays the part of an Internet archaeologist, I thought this was a common piece of knowledge five years ago, let alone today. Goes to show how incorrect assumptions about what we do (and do not) know can be.
Yeah, I just assume that today's 23-year-olds (or 13-year-olds) should all know this, and had the same experience as me, using notepad.exe to alter/test some html, which was grabbed from a "CD-ROM", that came with a "magazine", with something like "Netscape 1.0" (for Win 3.1) on it, in late 1994! What's The Frequency, Kenneth?
It's so obvious it hurts. The H in both HTML and HTTP is "hypertext". "ref" is clearly "reference". What kind of downer actually writes such a detailed article about this?!
A warm, fuzzy feeling knowing that we are the scientists the Jurassic Park cloners stood on the shoulders of to create the horrors that ensued in that trilogy.
So, I go to Google Groups to try to find the first mention of HREF on Usenet.
I have no idea how to do this. It takes four clicks just to get to the Google Groups page.
I see a search box. I look for something to give me the advanced search page. I can't find it. (I can see 2 buttons for settings.) So, I type HREF into the search bar and click seach. The drop down offers "in any group" or "groups named". First attempt I ignore both of those and just click the button.
Posts: 28200040, groups: 58
I need to narrow this down. I need the advanced search pane. It's still not there.
I arrange by date. LOL NO, that hasn't worked for literally years.
The advance options turn up if I'm in a group. They allow me to narrow the search down, but not to change the group I'm searching in.
I will donate $5 to the charity of your choice to the first person who tells me how the fuck I can perform the following search:
[HREF]
any Usenet group (or the news hierarchy)
Before 1998
Sorted by date
Bonus $5 if the instructions you give are in any Google documentation online.
It’s not a kludge. It’s a way to search websites that happen to have awful search (like Google Groups).
In most cases that works very well and it’s universally applicable. It’s a consistent search interface for (pretty much) any website there is. You can just throw Google at it and it works as good if not better than the websites’ own search (depending on how good that is), but it’s also a consistent interface you do not have to discover and re-learn for every website individually. Sometimes int
Kludge to me implies that it’s complicated to set up (it’s not; just “site:” in front of the website plus your search terms) or that is either not universally useable or at least requires a lot of manual adjustment every time you want to use it elsewhere (it’s not; universal and consistent search that works the same every time).
Sometimes a website’s own tools can be somewhat more comprehensive, sometimes they are so bad that switching over to Google is a necessity to find anything at all.
You're right. But in the context of the largest and best search engine in the world it's a bit kludgey to not be able to use their supplied search tools for one of their own products, because they broke it so much, and to have to rely upon a different product.
> consistent search that works the same every time
+ vs "" ?
> sometimes they are so bad that switching over to Google is a necessity to find anything at all.
Yes, Google search is a useful tool for websites that have hopeless search. I've used it for Wikipedia and other websites. It's baffling that Google's own product has search so broken that using a different Google product to search it is considered not-kludgey.
I like this kind of article because what seems obvious isn't always correct. It could have just as easily meant... anything... and seeing the process and resources different people use to dig into things is always pretty neat.
In real life language etymology, people joke (with some degree of seriousness) that if there is an explanation for a particular term or phrase, that is logical, reasonable, and has some historical corroboration, it is almost certainly wrong.
The evolution of language is a strange and unexpected process. I think this clearly applies to programming languages too. Surely more so for some languages than for others... but this seems to definitely be the case for web languages, since the ecosystem they evolve in is so chaotic.
It's funny. I half jokingly apply that maxim to a lot business decisions I see.
Someone will say "oh it makes sense that apartment a does this, because x y z" and I respond with "they may be doing something right, but theyre almost certainly not doing it hat way because it's right"
But yeah. It pays not assume the obvious answer is the right one
I'm surprised they didn't fix this sillyness in HTML5 when they were making up semantically-sensible tags for everything else.
A tag with
<link to="http://somepage" in="new window">
or something like that would make more sense as having the "article" inside the "body" of the document. Yes, I know link is already used in the head. You know what I mean, though.
I mean, the dual-use of a as both anchor and link has always been bizarre.
Most of the stuff HTML5 did is either backwards compatible in real-world HTML4 browsers or simply impossible; like, you've pretty much always been able to just drop down random tags that will default to acting sort of like a div. Changing how links work would be pretty severe.
> I mean, the dual-use of a as both anchor and link has always been bizarre.
In the original Ted Nelson "Xanadu" conception of hypertext (which inspired Berners-Lee and others at the time), all links were bi-directional.
We now use <a name=""> as a target and <a href=""> as a source but the original idea was that every link would be both a target and source – you'd link from one page to another then travel through the bi-directional link backwards and resume the original page.
Yeah, it didn't happen because it was grossly impractical didn't serve any real-world purpose. But that's why they share the same tag.
I just wonder how many people think that web is so wonderful invention. But they completely forget Gopher. There isn't so much difference between web and gopher after all. It's just a bit different linking method. After all, it's rare that invetions are relly great, most of those are just small gradual change over time. Still remember how much hype there was about hypertext back in old days. See: https://en.wikipedia.org/wiki/Gopher_%28protocol%29
Let's see - the HREF attribute that implements the basic principle of hypertext referencing in the Hypertext Markup Language which TBL invented as a file format to serve over his Hypertext Transfer Protocol... what could it possibly mean? I can see the confusion.
The answer might sound logical to you but all this time I thought the expansion was Hyperlink REFerence. My HTML education started in 1998 or 1999 from reading a PC Magazine series of articles on how to program in HTML. I don't remember how I got the wrong info and frankly didn't bother to care about this difference but I clicked this article with a smug face and reached the bottom feeling childishly embarrassed.
In the software world we often seem to end up with abbreviations which are ubiquitous but no longer strongly linked to their unabbreviated form. I often find this annoying and distracting. I try an avoid being the source of such annoyance for others. For example if I find myself using abbreviations in my source code, maybe as a prefix for example, I will always ensure the abbreviation is explained at least once in a place that won't be lost (in a header file for example).
An example from back in the day was the acronym (presumably) "Afx". Vast numbers of identifiers in the old MFC (Microsoft Foundation Classes) Windows API C++ wrapper used this prefix. For some reason not knowing what it meant caused me chronic stress.
stdafx.h is a Microsoft invention used to help speed up C++ compilation. Instead of parsing your 100000 lines of headers for every source file, you would do it once, and the compiler would dump its state to disk. Then, whenever you started another source file with `#include <stdafx.h>`, instead of repeating that work, it would load that state into memory and pick up compiling from there.
“One interesting quirk of MFC is the use of "Afx" as the prefix for many functions, macros and the standard precompiled header name "stdafx.h". During early development what became MFC was called "Application Framework Extensions" and abbreviated "Afx". The name Microsoft Foundation Classes (MFC) was adopted too late in the release cycle to change these references”
http://en.wikipedia.org/wiki/Microsoft_Foundation_Class_Libr...
Thanks. Actually this explanation reminds me that part of the frustration came from the feeling that Mfc would have been a much more logical prefix and wondering why oh why they hadn't used that instead.
I once asked a very experienced colleague in a Linux team to fix the modes on a file, and he wasn't sure what I meant.
Explaining the acronyms helps. It gives a feeling that things were actually designed, and allowed you to see the consistencies since those acronyms keep appearing in multiple places.
I'm old enough to have read Computer Lib ('74) when it came out, I have a first edition which I imported into the UK by sending some sort of international money order to a US bookshop. For some reason I also have a second edition.
51 comments
[ 3.0 ms ] story [ 51.3 ms ] threadBurners-Lee's talks are wonderful. My favorite is his talk "Hypertext and Our Collective Destiny": http://www.w3.org/Talks/9510_Bush/Talk.html
I never did use an IDE until lately, but Geany isn't much of an IDE compared to some other choices.
Everything goes back to Goldblum.
I have no idea how to do this. It takes four clicks just to get to the Google Groups page.
I see a search box. I look for something to give me the advanced search page. I can't find it. (I can see 2 buttons for settings.) So, I type HREF into the search bar and click seach. The drop down offers "in any group" or "groups named". First attempt I ignore both of those and just click the button.
Posts: 28200040, groups: 58
I need to narrow this down. I need the advanced search pane. It's still not there.
I arrange by date. LOL NO, that hasn't worked for literally years.
The advance options turn up if I'm in a group. They allow me to narrow the search down, but not to change the group I'm searching in.
I will donate $5 to the charity of your choice to the first person who tells me how the fuck I can perform the following search:
[HREF]
any Usenet group (or the news hierarchy)
Before 1998
Sorted by date
Bonus $5 if the instructions you give are in any Google documentation online.
Edit: sorry I bungled the conversion to the Julian format but you get the idea. You could also click around the 'search tools' sub-menu to do that.
Edit 2: It won't let you sort by date but you can just limit it to before 1994 :)
It only finds one message from 1993 (Dec 6), on comp.os.linux.development: https://groups.google.com/d/msg/comp.os.linux.development/Pk...
Who knows.
Oh my goodness.
Okay, so that sort of works. It's a bit of a kludge though.
In most cases that works very well and it’s universally applicable. It’s a consistent search interface for (pretty much) any website there is. You can just throw Google at it and it works as good if not better than the websites’ own search (depending on how good that is), but it’s also a consistent interface you do not have to discover and re-learn for every website individually. Sometimes int
Kludge to me implies that it’s complicated to set up (it’s not; just “site:” in front of the website plus your search terms) or that is either not universally useable or at least requires a lot of manual adjustment every time you want to use it elsewhere (it’s not; universal and consistent search that works the same every time).
Sometimes a website’s own tools can be somewhat more comprehensive, sometimes they are so bad that switching over to Google is a necessity to find anything at all.
> consistent search that works the same every time
+ vs "" ?
> sometimes they are so bad that switching over to Google is a necessity to find anything at all.
Yes, Google search is a useful tool for websites that have hopeless search. I've used it for Wikipedia and other websites. It's baffling that Google's own product has search so broken that using a different Google product to search it is considered not-kludgey.
The evolution of language is a strange and unexpected process. I think this clearly applies to programming languages too. Surely more so for some languages than for others... but this seems to definitely be the case for web languages, since the ecosystem they evolve in is so chaotic.
Someone will say "oh it makes sense that apartment a does this, because x y z" and I respond with "they may be doing something right, but theyre almost certainly not doing it hat way because it's right"
But yeah. It pays not assume the obvious answer is the right one
A tag with
or something like that would make more sense as having the "article" inside the "body" of the document. Yes, I know link is already used in the head. You know what I mean, though.I mean, the dual-use of a as both anchor and link has always been bizarre.
In the original Ted Nelson "Xanadu" conception of hypertext (which inspired Berners-Lee and others at the time), all links were bi-directional.
We now use <a name=""> as a target and <a href=""> as a source but the original idea was that every link would be both a target and source – you'd link from one page to another then travel through the bi-directional link backwards and resume the original page.
Yeah, it didn't happen because it was grossly impractical didn't serve any real-world purpose. But that's why they share the same tag.
EDIT: The one above, I meant. Thanks though, the downvote here too was well deserved. :D
An example from back in the day was the acronym (presumably) "Afx". Vast numbers of identifiers in the old MFC (Microsoft Foundation Classes) Windows API C++ wrapper used this prefix. For some reason not knowing what it meant caused me chronic stress.
TL;DR: it stands for Application Framework X (a group at MS responsible for the MFC, Microsoft Foundation Classes, library at some point)
Explaining the acronyms helps. It gives a feeling that things were actually designed, and allowed you to see the consistencies since those acronyms keep appearing in multiple places.
href => @,to,dest,target (others: http://json-ld.org/)
-------------
/etc => /config
/etc/rc.d => /config/start.d
[beyond POSIX]
Though I didn't know the A stands for Anchor, so there's that.
I'm old enough to have read Computer Lib ('74) when it came out, I have a first edition which I imported into the UK by sending some sort of international money order to a US bookshop. For some reason I also have a second edition.