The Great Works of Software
April 30, 2014 1:24 PM   Subscribe

Paul Ford on the five works of the software canon:

  • Microsoft Office: How do we communicate about work?
  • Photoshop: How do we create and manipulate images?
  • Pac-Man: How do we play?
  • Unix: How do we connect abstractions together to solve problems?
  • Emacs: How do we write programs that control computers?
posted by Chrysostom (93 comments total) 29 users marked this as a favorite
 
VI4EVA!
posted by grumpybear69 at 1:29 PM on April 30, 2014 [27 favorites]


What? No.
posted by sonic meat machine at 1:30 PM on April 30, 2014 [10 favorites]


Most of the "Western canon" is public domain so it would be better if this was all freeware. OpenOffice, Paint.NET, whatever game, whatever Linux, whatever language.
posted by turbid dahlia at 1:33 PM on April 30, 2014 [3 favorites]


No offense, but he left out what I consider humanity's masterpiece of the information age.
posted by mccarty.tim at 1:34 PM on April 30, 2014 [16 favorites]


The phone company gave birth to Unix. Now there is no phone company and Unix runs on your phone.
posted by ill3 at 1:37 PM on April 30, 2014 [33 favorites]


I think Tetris is probably a better contender for influential video game. It set a high bar for what a puzzle game can be, and of all the puzzle games I've played, I find myself coming back to Tetris more reliably than anything else. Plus, Tetris is simple and abstract, so it focuses on gameplay that's very engaging without attempting to tell a story or convey a realistic world, which I think shows something about video games that makes them distinct from books and movies.
posted by mccarty.tim at 1:39 PM on April 30, 2014 [8 favorites]


This is a very cool exercise, especially if you move past the "favorite software" angle and think about what a canon means.

1) Is it is the first version or the apogee of the form that makes the canon? Is it Visicalc or Excel that typifies productivity software?

2) Is it about elements or is it about complete experiences?

3) Is it about innovation? For example, this list of the 50 key video game innovations?

3) What about audiences? Is it for experts (Emacs) or is it for a general audience?

4) Is it about influence or about critical post-hoc views of importance?
posted by blahblahblah at 1:39 PM on April 30, 2014 [2 favorites]


  • Zombocom: How do we create a space in which all things are possible?
Yes, I will always laugh at zombocom jokes. Also any joke that involves something going faster than the speed of light.
posted by Mrs. Pterodactyl at 1:44 PM on April 30, 2014 [27 favorites]


You are standing in an open field west of a white house, with a boarded front door. There is a small mailbox here.

> █
posted by bonehead at 1:44 PM on April 30, 2014 [48 favorites]


humanity's masterpiece of the information age.

Yeah but I still need to learn Fractions 2.

4) Is it about influence or about critical post-hoc views of importance?

5) Is the cello part super tedious and repetitive and boring?
posted by kmz at 1:47 PM on April 30, 2014 [4 favorites]


Visicalc was the first spreadsheet program; Excel was the first Excel. Visicalc did spreadsheets. Excel does spreadsheets, but if you're the sort of person who works in Excel, it can be tortured to do a vast variety of other things, including databases, text editing, there might have been a point maybe where I used it to lay out an afghan pattern because I just needed something that was a grid of squares I could fill in different colors and I didn't have any graph paper? I think it's really the piece worthy of notice. However many features they've packed into it, most people still use Word like a text editor, occasionally one that does some basic formatting and mail merges.
posted by Sequence at 1:47 PM on April 30, 2014 [4 favorites]


The only software that matters.
posted by COBRA! at 1:49 PM on April 30, 2014 [10 favorites]


  • Microsoft Bob: How can we best relate to our computers on a personal level, kind of like in Her?
posted by mccarty.tim at 1:50 PM on April 30, 2014 [1 favorite]


There are other text editors, like Vim, and there are passionate debates over which is best.

Second only in Great Holy Wars Online to the right way to pronounce ".gif", though this is the best response to that I've ever seen and may bring peace in our time.
posted by fifteen schnitzengruben is my limit at 1:50 PM on April 30, 2014 [6 favorites]


Isn't it a little strange to use the language of aesthetics and talk in this literature-derived way about a "canon," when really these programs (apart from, arguably, Pac-Man) are basically functional objects, intended for use? I suppose there's no harm in assembling a "canon" of hammers, but that doesn't help you drive a nail. What Ford seems to want to do is really sociology, not aesthetics, and this vocabulary seems like the wrong tool for the job.
posted by RogerB at 1:50 PM on April 30, 2014 [3 favorites]


Lotus 1-2-3 was arguably the first Excel, and Wordperfect the first Word. That is both were the first fully-fledged versions of the products as we would recognize the feature sets today. The main differences, big differences, lie in the interfaces. However, the feature sets were largely the same.
posted by bonehead at 1:51 PM on April 30, 2014 [2 favorites]


For example, this list of the 50 key video game innovations?

I don't see procedural generation (Nethack, Elite, Diablo, Minecraft, Spelunky) on that list.
posted by straight at 1:52 PM on April 30, 2014


Pinnacle of software engineering. A list without this isn't even wrong.
posted by spikeleemajortomdickandharryconnickjrmints at 1:52 PM on April 30, 2014 [2 favorites]


Emacs! What? (And I use it not Vi)

How about

OS 360
Unix
Windows

More than half the developed world would hit those every day.

The next ones are harder.

SQL or IMS?
Netscape Navigator?
posted by sien at 1:52 PM on April 30, 2014


It's fun to think about this. Will we have software that is unknown and unloved at the time of its creation, only to find its place in the canon much later, like Moby Dick? Or the reverse (like, perhaps, John Dos Passos)?

I agree that Tetris has got to be at least as canonical as Pac-Man.

Though as a confirmed emacs user it pains me to say so, and though it delights me to see Ford relegate vi to the category, merely, of "other editors," I think one has to admit that, looking at it objectively, emacs and vi are equally canonical at this point.

Then there are all these little software subcultures with their own canons. There are the djbdns/qmail security people, the Plan 9/suckless.org/dwm people, etc., etc.
posted by enn at 1:53 PM on April 30, 2014


Wordperfect the first Word

Wordstar.
posted by MartinWisse at 1:56 PM on April 30, 2014 [17 favorites]


All I know is, Gorillas was history's greatest work of software for best-of-51 matches to accompany marathon afternoon drinking sessions in 1994.
posted by The Card Cheat at 2:03 PM on April 30, 2014 [3 favorites]


Isn't it a little strange to use the language of aesthetics and talk in this literature-derived way about a "canon," when really these programs (apart from, arguably, Pac-Man) are basically functional objects, intended for use? I suppose there's no harm in assembling a "canon" of hammers, but that doesn't help you drive a nail. What Ford seems to want to do is really sociology, not aesthetics, and this vocabulary seems like the wrong tool for the job.

People have aesthetic responses to all kinds of functional things: buildings, chairs, cars. (Maybe even hammers, I don't know.) I think there is absolutely an aesthetic quality to the experience of using software (sometimes intended by its creator, sometimes not), and people do respond on an aesthetic and emotional level to that.
posted by enn at 2:03 PM on April 30, 2014 [2 favorites]


uh -AutoCAD ?
posted by Abinadab at 2:04 PM on April 30, 2014 [4 favorites]


5) Is the cello part super tedious and repetitive and boring?

:golfclap:
posted by Sebmojo at 2:05 PM on April 30, 2014 [1 favorite]


I'd certainly replace MS Office w Visicalc, as that was the groundbreaker, and I'd think about DOS for Unix as far as popular impact, but thats certainly arguable and Unix is 'better.' PacMan? I dunno maybe Asteroids. Photoshop, hard to dispute in that genre. And Emacs? Not gonna go there. I'll go start a dino v synthetic thread in my bike forum instead.
posted by sfts2 at 2:07 PM on April 30, 2014


People have aesthetic responses to all kinds of functional things

Of course, but that doesn't mean that the tools and vocabulary of literary/art history are the right ones for a social/sociological history of functional things in use. That software has aesthetics is clear and uncontroversial, but doing software history as if it were software aesthetics is still a category mistake.
posted by RogerB at 2:08 PM on April 30, 2014


  • Netscape Navigator
  • The LAMP stack (Linux, Apache, MySQL, PHP)
The Internet is important.
posted by graymouser at 2:10 PM on April 30, 2014 [3 favorites]


cc
posted by Poldo at 2:11 PM on April 30, 2014


You could argue that Office being used to create other works is not dissimilar from the way the Bible and Greek myths are appropriated/interpreted to create subsequent works.
posted by 99_ at 2:12 PM on April 30, 2014


A software canon should be measured by the source code. This is a tool box not a canon.

As regards the noteworthy tools in that history of tools, DOS did not change things the way Unix did. Microsoft was significant, but not DOS.
posted by idiopath at 2:19 PM on April 30, 2014 [2 favorites]


WordStar, like VisiCalc, wasn't as full featured though: it had it's own weird control keys, it didn't reflow paragraphs properly, and so on.

It was, in many respects, more of a text editor than what we now think of as a word processor. In fact, I recall the WordStar command set mostly because they were embedded in the Borland Turbo Pascal and Turbo C IDEs.

That said, I agree with Fords pick of Office. The command line versions of WP (and WS), and 123 (and Visicalc) were ur-versions of ideas, important landmarks, but part of a developement which didn't get to stability until the early 90s. It's notable that these tools haven't changed much since.
posted by bonehead at 2:22 PM on April 30, 2014


enn: "I think one has to admit that, looking at it objectively, emacs and vi are equally canonical at this point."

objectively speaking, only one of those is part of the Single Unix Specification, and it ain't emacs. ;)
posted by namewithoutwords at 2:27 PM on April 30, 2014 [4 favorites]


Is there a reason we're using this gigantic, cumbersome header graphic?
posted by humboldt32 at 2:28 PM on April 30, 2014


I can't believe Windows Millennium Edition didn't make the cut. They fixed the Y2K bug so we wouldn't have to eat dogs in lieu of modern banking!
posted by oceanjesse at 2:35 PM on April 30, 2014 [2 favorites]


Everything on this list either is, could be, or should be a subset of *nix. There's your canon...
posted by jim in austin at 2:42 PM on April 30, 2014 [1 favorite]


Uh, where is "hello world"? I can't take this list seriously.
posted by Philosopher Dirtbike at 2:51 PM on April 30, 2014 [1 favorite]


What about gcc and make?
posted by humanfont at 2:52 PM on April 30, 2014 [7 favorites]


Everything on that list is stand-alone! He missed anything to do with how we use this thing called a network.
posted by Dashy at 2:58 PM on April 30, 2014 [1 favorite]


WordStarHipHop
posted by Small Dollar at 3:01 PM on April 30, 2014 [2 favorites]


To expand my earlier comment (I was waiting for a build!), I wanted to come back and explain my opinion on this. To my mind, it cannot be a "great work" if the only way people can interact with the software is to buy it and use it as an end user. It needs to have significance to the later development of software, the way great literature is significant for the literature that follows it. It also helps if it's been analyzed in depth. In that vein:
  • TeX. Covered in Computers and Typesetting, volumes A and B, by its author, Don Knuth.
  • Unix. Covered in numerous books, notably The Unix Programming Environment by Kernighan and Pike.
  • C. Covered in The C Programming Language, by Kernighan and Ritchie.
  • HTTP. A protocol, not an implementation, but nevertheless. RFC 2616.
  • Lisp. Originally proposed around 1958, later covered in a number of great books. The Structure and Interpretation of Computer Programs, notably.
  • vi. No great books, but vi influenced the entire development of the console display (via libcurses).
  • emacs. Many books, among them Stallman's. Most significant for its extensibility.
  • Quake. A performant, true 3D virtual environment. No book, but the source code is available.
User-space stuff, to me, is interesting in some ways; but most of the techniques that it uses are either proprietary and secret or were invented somewhere in academia long before they were used. Cramming user-features into software isn't very interesting. It's harder to keep software simple than to make it complex; and so the most interesting software is often deceptively simple.
posted by sonic meat machine at 3:05 PM on April 30, 2014 [23 favorites]


This kind of feels like people arguing about whether George Harrison or Randy Rhoads was a better guitarist.
posted by thelonius at 3:22 PM on April 30, 2014 [1 favorite]


The only software that I agree with on this list is Photoshop. It was a game changer for image processing and manipulation. It has only become more of a standard over time. And despite many attempts to clone it, there hasn't been anything nearly as good to replace it.

That said, here is my list:
  • The Web Browser: How do we communicate about work?
  • Photoshop: How do we create and manipulate images?
  • Super Mario Brothers: How do we play?
  • Mac OS GUI: How do we connect abstractions together to solve problems?
  • BASIC: How do we write programs that control computers?
The web browser is the gateway to the internet for most people. Even in this post-PC world, with apps and social networking, I would argue that a Facebook, or Twitter or other social networking app on a phone or tablet is just a specialized web browser dedicated to a specific task (and a loose attempt at portal lock-in).

Super Mario Brothers, in my opinion, is a more relatable and influential game than Pacman. I think it is more recognizable and has had more of an impact on how games are created and designed. Mario Brothers itself probably reached its pinnacle on the SNES with Super Mario World, but the influence seeped into the DNA of game programming and gameplay. Nearly anyone who has played video games understands the mechanics of platformers. It was Super Mario Brothers and its successors that ingrained that to the world.

Yes, there was Xerox's attempts at GUI (amongst others), but it was the Mac OS GUI that got the formula right. It became the paradigm to interact with personal computers for decades and is still at use in its evolved form in the phones and tablets we use today.

BASIC isn't used as much today as in the past, but it was a HUGE influence on many of the innovators of programming history. Bill Gates laid the foundation of the Microsoft empire with his version of BASIC. A ton of young programmers were able to tell their computers to do something through BASIC's simple to understand commands. It was that introduction that allowed them to conceptualize and realize how to control computers and graduate to more robust languages for "real" programming. It would be great if there was as an equally ubiquitous and accessible language on our modern devices, though I think the argument could be made that HTML and its successor variations might be in the same spirit (if not a bit more limited).

I would be interested to see other MeFi member lists.posted by bionic.junkie at 3:23 PM on April 30, 2014 [2 favorites]


I don't think that "first" comes into it. When people talk about canon in other fields, it's not what was the first violin concerto is it? If anything I probably thing that Office is the one thing that most clearly deserves to be there.

Even Photoshop which I initially thought clearly deserved its place is really just showing my, and the author's, biases. Why not Autocad or Solidworks or some other ubiquitous bit of software from another industry. Why would "creating and manipulating images" be more important that sound or music or 1000 things from industry.

Even the Unix entry: "It runs on your Mac!" Whose Mac? Desktop computers are still 90% Windows.
posted by markr at 3:34 PM on April 30, 2014 [1 favorite]


We're such damn narcissists in tech, I swear to god. How could pacman possibly be more important than Unicode?
posted by oceanjesse at 3:37 PM on April 30, 2014 [2 favorites]


Even the Unix entry: "It runs on your Mac!" Whose Mac? Desktop computers are still 90% Windows.

It runs on your iPhone and Android phone (Unix and Linux respectively).
posted by bionic.junkie at 3:44 PM on April 30, 2014 [1 favorite]


Not to badmouth the only one of these that I have had code in, but Microsoft Office is not a program, it's a bundle containing whatever disparate software they decided to shove in there for that version. PowerPoint was acquired from Forethought, some versions have included Internet Explorer, and at one point the premium Mac version included Virtual PC (written by Connectix). This is like saying your favorite food is "lunch".
posted by w0mbat at 3:47 PM on April 30, 2014


Scripsit 4 lyfe!
posted by jadepearl at 3:57 PM on April 30, 2014


By that same argument Unix would be excluded. Since it is the very definition of thousands of separate programs. (Not that I'm saying Unix shouldn't be there, it definitely belongs on the list).
posted by markr at 3:59 PM on April 30, 2014


> Most of the "Western canon" is public domain so it would be better if this was all freeware. [...] Linux

I like the idea of public domain projects, but something derived from the original Bell Labs Unix like a BSD or illumos strikes me as better fitting the historical attribute.
posted by morganw at 4:00 PM on April 30, 2014


For certain areas, like compilers or games, you really are building off design ideas that were invented by the "classics" and just need minor revisions to make them your own. But most software is poorly defined and messy and doesn't clearly fit in with something that has been done in the past.
posted by miyabo at 4:01 PM on April 30, 2014


How could pacman possibly be more important than Unicode?

I don't think this is about the idea that what didn't make the list has some kind of objectively smaller impact on the world -- it's about naming 5 thing that have had a visible cultural impact, and what they say about humans and culture in their sphere.

So, PacMan is certainly orders of magnitude less important as a practical tool, it casts a long shadow, and to examine it is to look at the fact that we'll spend effort to use new tech to create new forms of play and entertainment. There are arguably other games you could pick as a jumping off point for that discussion, but Pac Man is a defensible choice.
posted by weston at 4:08 PM on April 30, 2014


Ford states his criteria:

So I set myself the task of picking five great works of software. The criteria were simple: How long had it been around? Did people directly interact with it every day? Did people use it to do something meaningful?
posted by thelonius at 4:12 PM on April 30, 2014


LaTeX and sendmail are probably the closest to software canon we're going to get. They're both rock solid, tested to high hell, barely (if) updated, and pillars of the software community. The rest, though? Microsoft Office? No way. Not canon.
posted by spiderskull at 4:25 PM on April 30, 2014 [2 favorites]


No M.U.L.E. == F.A.I.L.
posted by meehawl at 4:27 PM on April 30, 2014 [3 favorites]


cubase for the atari ST or cakewalk for DOS, with honorable mention going to the ultimate soundtracker for amiga and rebirth 338

all had a massive impact on today's computer music software
posted by pyramid termite at 4:28 PM on April 30, 2014


Question: how many of those five were chosen specifically for the controversy they would generate?
posted by Tell Me No Lies at 4:30 PM on April 30, 2014


There are the djbdns/qmail security people

Yo! Represent!
posted by mikelieman at 4:32 PM on April 30, 2014


Again, the C compiler. It's been around since forever, it's still #1 in popularity, and most if not all of the other examples of canon people have given are written with one.
posted by Poldo at 4:34 PM on April 30, 2014 [7 favorites]


still #1 in popularity

Holy hell, PHP is third?
posted by thelonius at 4:42 PM on April 30, 2014 [1 favorite]


Which C compiler though? The selections weren't "office packages, image editors, games" etc. They were explicit pieces of software. Certainly you could make an argument for gcc.
posted by markr at 4:45 PM on April 30, 2014 [1 favorite]


PHP is third?

as mentioned in Revelations, a sign of the coming apocalypse
posted by me & my monkey at 4:45 PM on April 30, 2014 [3 favorites]


pyramid termite: MUSIC1 originated the underlying structure shared by nearly everything that followed, from csound to max/msp to cubase to ableton live to frooty loops, they all use some trivial transformation of the MUSIC1 design with more or less flashy pictures on top.
posted by idiopath at 4:58 PM on April 30, 2014 [1 favorite]


that's an interesting point, idiopath, but just how many people used it? - nothing against the awesome efforts of max mathews, but frankly, entering in data by punch card isn't my idea of a fun or efficient workflow - (in fact, it turned me off computers for a long, long time ...)

actually, i'm not totally certain amiga tracker software owes much of anything to MUSIC1
posted by pyramid termite at 5:12 PM on April 30, 2014


As others have pointed out, this is the software canon only for people who know nothing about software or computers.
posted by GuyZero at 5:13 PM on April 30, 2014 [1 favorite]


pyramid termite: punch cards are just an antiquated UI, having nothing to do with the audio engine.

MUSIC1 is the origin of audio processing with separate update rates for time varying controls vs. audio signals, with a fixed size audio buffer vector per scalar control input. The list of usable, realtime capable audio software that does not borrow this design is very very short.
posted by idiopath at 5:17 PM on April 30, 2014


I haven't heard of MUSIC1 and I've failed at searching for it effectively. Can someone drop me a link?
posted by jepler at 5:17 PM on April 30, 2014


MUSIC is the appropriate name, sorry.
posted by idiopath at 5:18 PM on April 30, 2014


How I expected the rest of the list to go after seeing Microsoft Office at the top:

sendmail: A Turing-complete mail server.
COBOL: Common Business Oriented Language.
INTERCAL: Another deserving programming language.
CHRISTMA EXEC: Just type CHRISTMAS from cms.
Comet Cursor: Change the appearance of your Microsoft Windows cursor.
posted by sfenders at 5:21 PM on April 30, 2014 [2 favorites]


"Microsoft Office" or "Photoshop" is not an explicit piece of software unless you put a version number beside it and Unix is a specification, not a piece of software. As for the idea that the list is for "people who know nothing about software", I'd be surprised to learn these people know anything about Unix, let alone emacs.
posted by Poldo at 5:27 PM on April 30, 2014


Also I nominate Perl. Perl is like Revalations. Weird, impenetrable and probably a bunch of gibberish. As it is written "And then I saw Perl7 and behold a new Jerusalem. A choir sang regexes from the heavens...."
posted by humanfont at 5:28 PM on April 30, 2014 [3 favorites]


pyramid termite: punch cards are just an antiquated UI, having nothing to do with the audio engine.

but it has everything to do with the availablity and "popularity" of software in that era - which is not the era of the canon suggested above

besides

(gearslutz) it's not analog, is it? (/gearslutz)
posted by pyramid termite at 5:30 PM on April 30, 2014


Aha, thanks idiopath
posted by jepler at 5:31 PM on April 30, 2014


The interesting thing to me is how much the basic design of most software has changed since that time, but somehow MUSIC came up with something that still has yet to be significantly changed. That is one of the things that canon means to me.
posted by idiopath at 5:37 PM on April 30, 2014


This is like saying your favorite food is "lunch".

If only one company made lunch for 12 years.

Its more than the sum of its parts. Its the idea that these products are functionally related in such a way that you would want those products in there for a basic office setting. And then adding pretty functional interoperability. If it was a fucking failure it would have collapsed, attached to Windows or not.
posted by Ironmouth at 5:41 PM on April 30, 2014 [1 favorite]


MUSIC1 originated the underlying structure shared by nearly everything that followed, from csound to max/msp to cubase to ableton live to frooty loops, they all use some trivial transformation of the MUSIC1 design with more or less flashy pictures on top.

Hahahahahaha. I get your point, idiopath, but that's a bit like saying that Beethoven's 5th is a trivial transformation of the chromatic scale.
posted by speicus at 5:46 PM on April 30, 2014


It doesn't fit into the criteria mentioned in this article, but I feel that many pieces of malware could be included for their impact, either in the individual or the collective. I particularly think that the Stuxnet malware could count as being among the most game-changing software packages ever put together.
posted by all the versus at 5:55 PM on April 30, 2014


Explaining why GNU’s Not Unix would ruin your afternoon; just assume it has to do with powerful feelings felt by programmers in the early 1980s.

But that's kind of missing the whole point behind Emacs and why it is truly great software. The extensibility of the macro language addresses the core philosophy of freedom precisely as espoused by GNU. This author missed the whole point, discounting the cultural aspects/critique of software. Same goes for his explanation of Office.
posted by polymodus at 5:56 PM on April 30, 2014 [1 favorite]


I think that this exercise would greatly benefit from having categories - canonical office productivity software, canonical programming languages, canonical games, etc.
posted by all the versus at 6:04 PM on April 30, 2014


lasnex
posted by FuturisticDragon at 7:13 PM on April 30, 2014


TeX is unique in that the main source code branch started in 1978 is still maintained by its original author, in its original programming language, despite huge advances in technology (Unicode, PDFs, etc) and is still in widespread use. Can't say that about C compilers, UNIXes -- or even emacs, which has been forked.

Knuth's ambition was to write software as if it were literature, and I daresay he succeeded.
posted by RobotVoodooPower at 7:30 PM on April 30, 2014 [5 favorites]


How about a canon of practical concepts, I.e. patterns? MVC, decorators, inheritance, ... As long as the central criterion is influencing succeeding generations of programs, might as well focus on the reusable concepts themselves.
posted by deathpanels at 9:29 PM on April 30, 2014


#include <unistd.h>

int main(void)
{
while(1) fork();
}
posted by simra at 9:37 PM on April 30, 2014


I submit that the masterpiece hidden in MS Office is OneNote. Word/Excel/Powerpoint are all important pieces of software for their own reasons, but IMO nothing captures the essence of usability better than OneNote.
posted by simra at 9:44 PM on April 30, 2014


Its more than the sum of its parts. Its the idea that these products are functionally related in such a way that you would want those products in there for a basic office setting.

What's more, the way any organisation uses Office is a good clue for their structure and mentality. The three basic components of Office are Word, Excel and Powerpoint. As any one can tell you who has worked for multiple clients, whether they prefer Word over Excel or Powerpoint over Word, matters a lot.

Powerpoint: ideas are sketched out, mainly communicated through pictures and much of the work has to be done by the user/reader. Can be pie in the sky, can be agile in the best sense of the word.

Excel: numbers orientated, precise, often to the point of foolishness, rigid, prepared, controlling.

Word: language orientated, prolix, abstract or woolly, room for initiative as long as it's documented later.

(Outlook: consensus seeking, people orientated rather than process, ass covering, sometimes too chatty; Communicator: even more so; Access: run like hell)
posted by MartinWisse at 12:28 AM on May 1, 2014 [3 favorites]


What about Access?
posted by thelonius at 4:13 AM on May 1, 2014


Oh, sorry, I missed that - "run like hell".

Access is interesting (Office-wise) because there is no clear real world metaphor for a database. The other applications in Office have one. Word: a blank sheet of paper. Excel: graph paper (or an old paper accountant's spreadsheet) Powerpoint: slide carousel. Access.....a filing cabinet? That's kind of the opposite of a normalized database, since, in a filing cabinet, you make multiple copies of things, and store them in different places.

I think that is the non-technical reason why people's Access databases are usually so terrible: they use it to imitate a paper filing system (the technical reasons all come down to it being a file based, non client/server database, that doesn't handle multiple connections well, and which doesn't really have transactions).
posted by thelonius at 4:27 AM on May 1, 2014 [1 favorite]


What about Access?

The message Access sends is, "'Kind of' is good enough for us. We're so half-assed that we saw our chairs in two and buy one for every other employee."
posted by wenestvedt at 9:31 AM on May 1, 2014


Question: how many of those five were chosen specifically for the controversy they would generate?

This is controversial? What a world we live in!
posted by juiceCake at 10:40 AM on May 1, 2014


This is controversial? What a world we live in!

There is no religious conflict that comes close to emacs vs. vi.
posted by Tell Me No Lies at 10:58 AM on May 1, 2014 [1 favorite]


the internet was written on 'Next' plus Apple and shit...
posted by judson at 11:29 AM on May 1, 2014


Interesting article for stimulating discussion but no more answerable than "What are the 5 most important pieces of music?" Within Office I would single out Excel. I use mostly wikis for documents now - I probably create like 2 Word docs a year and I've never created a PDF other than via export. But nothing has really replaced Excel for number crunching and graphing data. We have an intranet with like 200 fancy reports we've created for users but they all have a "Download source data into Excel" button because it's so much easier for people to just get what they want using Excel. In fact when I get asked for a new report I ask people if they will ever want to customize the columns, work with the data, etc. and if so we just give them an Excel download link and no web report at all - why bother?
posted by freecellwizard at 12:00 PM on May 1, 2014 [1 favorite]


I don't think the field of literature has been especially well-served by having a canon and innumerable arguments about which works belong in it and which do not, so I'm dubious of the value of creating one for software.
posted by whir at 1:29 PM on May 1, 2014 [2 favorites]


« Older There is nothing.   |   Font samples Newer »


This thread has been archived and is closed to new comments