Slashdot Mirror
Command Line for the Web
flood6 writes "SearchEngineWatch offers a look at a new method of interacting with the Internet, YubNub. This 'social command line for the web' lets users create commands that interact with websites. Currently, most of the commands apply to search, but new commands could work with any site that accepts variables passed with HTML's GET command. For example, iap moon would search the Internet Archive for all media related to 'moon'."
I like the concept, however the premise is the flaw. What I mean by this is that the social concept of letting people add their own commands is the biggest draw, but the biggest flaw. Doing a quick ls on the site reveals such choice commands on the first page:
p Created 2005-06-20 16:21 - Description - Nominate - www.pogo.com --does nothing
tiki Created 2005-06-20 15:10 - Description - Nominate - http://www.tehanitrading.com/ --takes you to somone's website only
And then of course you have the sophisticated ones:
fuckthefucker Created 2005-06-20 12:55 - Description - Nominate - http - does nothing
And if you try to create a legitimate command and something like "p" is already taken you are out of luck. This is why you will need someone to moderate it from time to time (this is sort of on the to do list by flagging spam commands, but it looks like they also have a potential problem with bot submissions as well). But, once you do this then you get the problem of my command is better than your command. It would be nice to see an individual implementation whereby you could store your own commands and could "share" them with others.
Quality Hosting e3 Servers
John Gruber wrote a great essay on this called "The Location Field is the New Command Line". As he put it, "Web apps are just so damned easy to use ... It's all about the fact that you just type the URL and there's your email."
Wouldn't that also include all Internet Archive Project media featuring someone mooning you?
I'm scared.
Maybe they meant HTTP?
I haven't RTFA, but the example in the summary sounds pretty much like quick searches anyway.
Oh no... it's the future.
# p2p Blahsong.mp3
This would seem to have more worth if it was done with a plugin/toolbar. Right now, it requires that I switch to my address bar, type in yubnub.org, wait for it to load, switch to the form input, and type in my command.
Much better if one could skip straight to the command part.
-dave
http://millionnumbers.com/ - own the number of your dreams
I know, bad geek humor, but it does kinda seem like a throwback to gopher.
A goal is a dream with a deadline
You can never have too much Ruby on Rails, ya know?
Now we can have old style Sierra adventures combined with the joy of Internet graphics...
...
Whee!
Command: GET SUBMIT BUTTON
"Ok! You got it."
Command: USE SUBMIT BUTTON ON POST
"You die."
blast, I was never very good at these games.
MoM++ - A Classic Expanded - [Master of Magic 1.5]
http://mompp.sourceforge.net/
Please stop trying to make the web do things that there are much better-suited alternatives for. Wouldn't this be so much more usable and efficient as a perl or similar script/library, that you could use not only to get things manually but as part of a greater program?
I am trolling
Like another poster has mentioned, this does indeed sound like an extension of the "location field" as the command line. However, I can imagine some pretty interesting inventions, if you could take the output of one of these commands and, like the typical command line, pipe them into another program or web application. Even better would be web applications that could recognize the piped output for what it was.
Imagine the following, running as a cronjob:
"del.icio.us --check | gmail me@home.com 'Cool Links'"
Or something like that. I lack imagination - I'm sure others could think of interesting ideas as well.
concrete5: a cms made for marketing, but strong enough for geeks.
They probably meant HTTP GET, rather than HTML GET.
(Hyper Text Transport Protocol)
(Hyper Text Markup Language)
See the difference?
I haven't RTFA either, but am I the only one who sees this being a terrific tool for all kinds of pandemonium on shoddy (i.e. amateur) Web sites?
It's not a lie. It's the truth with lossy compression.
It needs some sort of namespaces. An example:
org.archive.search moon
To do the moon search in the example. This will solve the problem of people fighting over command names.
But it still isn't going to work because the number of possible commands is going to grow to such a point that no one will remember what command to use! When I use bash and tab complete on "x", I get 119 possible completions. With this command line, type "com." and you get thousands of possible completions. Maybe it will be sucessful if it never catches on and doesn't have to scale...
You can see some more of Jon's gadgets here.
SEO Firefox Extension
I just saw this on the page: http://www.yubnub.org/documentation/describe_insta llation/
-dave
http://millionnumbers.com/ - own the number of your dreams
Maybe we could expand it to all parts in the Internet. Then people could do cool stuff like check their mail at the prompt, read newsgroups at the prompt, maybe even read each others blogs at the prompt. Hey, we could call it "telnet" or something really catchy. This is just crazy enough to work, guys! Who's with me?
I'm starting my CLI-accessable blog right away. I'm gonna call it the "Finger".
Ain't progress grand?
-Tom
Not to be pedantic, but the GET request belongs to HTTP, not to HTML. HTML is a fixed file format that can be transported across any medium, including NFS/FTP/SMB. HTTP is a transport protocol that allows some amount of negotiation between file and server, and it's possible to use completely separately from HTML in some cases.
no i am not kidding
Lets see if the framework lives up to the hype.
Will it survive a slashdotting ?
www.*> rm -rf
Experiment!
I want a REGEX web search. Can't think of an example off the top of my head, but there have been many times that I've wanted one for extra control.
WeRelate.org - wiki-based genealogy
The best version of this concept that I saw was XMLTerm. It's a graphical CLI, in a web page. It's ideally suited to be a CLI for websites, as well as a website implementing a CLI. It was ahead of its time, and went down the drain when the bubble popped. On purely "zeitgeist" problems - it appears to be technically sound. If there's interest, and effort, in this kind of app now, it will be much better to pick up this orphaned OSS project, than to reinvent it.
--
make install -not war
c:> GET ALL P0RN
Error reading device 'Signature'. (A)bort, (R)etry, (F)ail?
No, the Location Field would only be the new CL, if people bragged about using it, and whined that people who DON'T use it "don't understand how their computers work, did you know some users can barely write a simple device driver!"
Freedom: "I won't!"
Is "social" the new overused buzzword of the week?
Now that that's been taken care of, I'd like to say that this is an interesting idea. As a sysadmin I spend a lot of time in a terminal and though this has not been integrated into the terminal yet, the idea of being able to run a web search through a command line interface makes me happy.
I can just imagine:
Now what would be even cooler: You know the console from all the first person shooter games like Quake and Half-Life. Really simple, just hit Tilda and it slides down the top third of the screen. The quake 3 one even has some nifty open GL moving background. Anyway, my point is, how come to get a console in Windows I have to hit WIN+R, "CMD", [Enter]. Not fair. I've searched near and far and havent found any such console for Windows or Linux.
If a fellow slashdotter could point me in the right direction I would greatly apreciate it. I know this is a good idea, and I know I'm not the only one who would love such a program. It would be so nice to simply hit ~, run your ipconfig command, then ~ again and go back to what you were doing.
Aardwolf
Im dreaming ofa big bndwdth, That can resist the
if you do an ls of available commands, a lot of the commands are advertisements for blogs.
$ php Blahsong.mp3
i never p2p as root
ps, shove the okmmrfj up your ass taco
http://yubnub.org/example/tts?args=i+don't+have+an y+speakers+you+insensitive+clod
The only advantage of this is that it's collaborative... and as other users have pointed out, quite a lot of the commands on the site are crap and need moderation of some sort.
;)
But your browser has this built in already.
-If you use IE, you can set up these type of search shortcuts using TweakUI.
-If you use Firefox/Mozilla, you can create bookmarks that implement these sort of shortcuts. There's some examples in your bookmarks menu when you first install Firefox, just look at those for how to do it.
In both cases, after creating them, you can just type "shortcut search terms" in the address to make it do that "search". Doesn't have to be a search of course, it can be any kind of HTTP GET that you want. I have several defined..
-g for quick googling
-imdb for movie lookups
-imdbq for movie quote lookups
-snpp to search the simpsons archives..
-And so on.
- Give a man a fire and he's warm for a day, but set him on fire and he's warm for the rest of his life.
Yubnub was a railsday entry. Pretty neat what can be done in mostly one day.
It will be interesting how rails handles under the slashdot load.
Producing satire is kind of hopeless because of the literacy rate of the American public. - Frank Zappa
Yub nub, eee chop yub nub,
toe meet toe pee chee keene, g'noop dock fling oh ah.
Yah wah, eee chop yah wah,
toe meet toe pee chee keene, g'noop dock fling oh ah
Coat ee chah tu yub nub,
Coat ee chah tu yah wah,
Coat ee chah tu glo wah.
allay loo ta nuv
Glo wah, eee chop glo wah, ya glo wah pee chu nee foam,
ah toot dee awe goon daa.
Coat ee cha tu goo (Yub nub!)
coat ee cha tu doo (Yah wah!)
coat ee cha tu too (ya chaa!)
allay loo tu nuv (3 times)
Glo wah, eee chop glo wah.
Ya glow wah pee chu nee foam,
ah toot dee awe goon daa
allay loo tu nuv.
of what to call the commands and what they should perform when moz/firefox already allows the same thing via bookmark keywords? maybe i'm misunderstanding the point but this just seems like an interesting idea that is doomed to be a disaster. i have a hard time seeing where it could be useful. by the time i go to this site, look for a command, and figure out just how to use it, i could have easily made my own powerful bookmark with a name *i* invented for it, which makes sense to me and is three letters long (it won't take long for short names to be impossible to score in this thing).
It sounds like something an Ewok would say.
All the worlds indeed a
Lynx, archie, veronica, gopher, ping, traceroute, whois, nslookup... sound familiar to anyone else?
"fuckthefucker
Created 2005-06-20 12:55 - Description - Nominate - http"
Wow I don't remember seeing that command anywhere before. Maybe having some kind of moderation system to filter out the crap commands and ban websites for abuse? Really though I don't see my Mom using a command line - typing a word, then pointing and clicking is about the total amount of effort the average joe/janes is willing to commit to computing.
They need a 'filter the stupid stuff' applied real quick or this will be relegated to the dustbin of the web. Really, the balance between 'needs clever users' vs results is too skewed.
Does kinda make sense, I just saw a 'VC blog' with the question: 'is the bubble back'.
Dream on.
Buy tulips.
I can type
l for how to use moz keywords)
imdb Terminator
into my mozilla address bar and it does an search on imdb.org for terminator. or
dict antidisestablishmentarianism
and it searches on dictionary.com for "antidisestablishmentarianism"
the only advantage to this that I can see is that it has addresses that I might not have known about. But there are probably better ways of going about this than making a new web page. Perhaps a blog or plugin system for picking which ones you want mozilla to use.
(see http://www.mozilla.org/docs/end-user/keywords.htm
Interesting, I just made one for GRC port probing: http://www.yubnub.org/kernel/man?args=grcscan
Slashdot needs Geekcode | Can anyone recommend any good SCIFI? My tastes: Foundation, Startide Rising, CITY, Ringworld,
I can't find a single practical use for this. Nor do I find it very intuitive at all. Call me *old fashioned* but things were fine the way they were.
If you need me I'll be over at this sane website...
google.com
"why don't you just slip into something more comfortable...like a coma!"
I thought they got rid of the YubNub music at the end of Jedi. Thank goodness too!
In FF, you can have bookmarks like this: http://livejournal.com/~%25s/ Give it a keyword (say, 'lj'), type in the URL line: lj someusername and someusername's livejournal opens. This is the simplest example. I have several more sophisticated; this mostly obviates the FF search box. And all this with a *very* simple syntax that only allows substitution of one string. Imagine something a bit more powerful in that place.
Computers make very fast, very accurate mistakes
I suppose it shouldn't be a surprise that this is going on. Many have long preferred the CLI for interacting with applications. The web is undergoing a change from a medium for simply displaying information and become more of a protocol for client/server web applications. Is it any wonder that as the universe of web apps grows, that people want CLI utilities to communicate with them?
It's nothing new. We've been running finger, whois, nslookup and so on from the CLI for ages; these are utilities that could just as easily run over the web. (As opposed to say, IRC, which is hurt by the lack of statefulness)
I suppose this is yet another reason why web applications should carefully separate their presentation from other logic, since they may be called upon to present data to entirely different clients in the future if this became more common.
Comment removed based on user account deletion
Oh, and you can also use the address bar to ssh for your files (fish:) or get the images of your camera (digikam:). Bow to the power of the KIOslave framework!
Any sufficiently advanced libertarian utopia is indistinguishable from government.
I think its worth mentioning that YubNub was created for the RailsDay. A one day Ruby on Rails coding contest which was featured in the developer section of slashdot a week or so ago.
Its not the pinnacle of complexity but its still amazing what you can do in just a day. Hats off the the programmer.
This is called lynx. I did this ten years ago. (Time flies)
Try this
alias YubNub lynx
Stephan
http://stephan.sugarmotor.org
"i get me porn"
Fusica.com has a similar feature under expert search except that it is not collaborative.
FYI, there is a firefox search engine plugin for this too. http://mycroft.mozdev.org/download.html?name=yubnu b&submitform=Find+search+plugins/
Unfortunately the slashdot command only searches slashdot. It would be much more accurate if this command would bring nearly any web server to a screeching halt.
This is pretty much useless if you use Mozilla/Firefox, where using Custom Keywords, your Navigation bar becomes your command line.
I had been collecting my favorite keywords on my Custom Keyword Wiki
The wiki is now closed due to a lost battle against spammers, but as far as I have seen this is the biggest list of custom keywords available, so if anyone wants to pick up the project they are more than welcome. Email to awormus(a)gmail.com.
Just bookmark any page that is from a GET query. E.g. an IMDB search for "Batman Begins" gives you this:
b egins
http://www.imdb.com/Find?select=All&for=batman%20
Modify the properties of the bookmark, replacing "batman%20begins" with %s. This is a placeholder.
Give the bookmark a keyword, such as "imdb."
Now you can type "imdb X" in the url bar in firefox, hit enter, and it will do a search for whatever you enter for "X." Much easier than using yubnub.org.
I have bookmark searches setup for all kinds of stuff. Whois, nslookup, tracert, imdb, dictionary, gg (google groups), gi (google image), gm (google maps), yyp (Yahoo Yellow Pages), the list goes on and on. Any URL that accepts query words will work for you.
Ironically, the word ironically is often used incorrectly.
they should add a command where you can slashdot a site at will. Let the melting of servers begin!!
Did anyone do a ls on yubnub.org and realise that "vagsmell" was a command?..
I gave the bat commader a high five.
<BODY BGCOLOR="#000000" text="#FFFFFF">
<FONT face="sans-serif">C:\><BLINK>_</BLINK></FONT>
</BODY>
</HTML>
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
I am going to expand on your CLI-accessible blog to only provide the blogs from the middle of the page.
I will call it the "Middle Finger".
You can't handle the truth.
We already have short GETs -- add a firefox bookmark with a %s in the url, give it a keyword, then see as you can type "keyword arg" into the address bar and have it expanded. Now we just need web services to come up with sensible arg names~
I mod down anyone who says "I will be modded down for this", regardless of the rest of their comment
The K desktop environment has things called "KIOslaves" which recognize certain pseudo-protocols on the Konqueror location bar (and other places).
Internet examples include "deb:" for Debian package searches, "rf:" for rpmfind, "gg:" for Google, "ggl:" for Google-I-feel-lucky, and "rfc:" for getting RFC text from the IETF website, and "wp:" for Wikipedia. There are lots of these.
Non-internet examples include "man:" for viewing man pages, and "info:" for viewing those otherwise horrible GNU info pages.
2*3*3*3*3*11*251
- Give a man a fire and he's warm for a day, but set him on fire and he's warm for the rest of his life.
And give you a virus if you're not careful.
http://www.dqsd.net/
Sorry, I could not resist.
:)
Just try the "recursive" command
Engage!
Yub Nub means "Freedom" in Ewokese.
yes.
fucking ewoks.
yes.
I'm lame.
(former Official Star Wars Fan Club member 1983-1985)
To get a linux console in the same style as the quake console, hit ctrl+alt+f1 (or most of the other function keys). You have to log in the first time you use it.
o ad"
For windows you can put an icon in the quick launch bar instead of going through the run menu.
$ lynx "http://www.google.com/search?q=firefox+ftp+downl
or maybe links instead of lynx depending on your system. You can use a script to make that a little easier to type, and you can download in lynx so you don't need to use a seperate ftp client.
p.s. Sysadmin? If you can get a job without knowing this stuff then I'm set for life! Yay!
High School/Jr.High 10 PRINT "HELLO WORLD" 20 END First year in College program Hello(input, output) begin writeln('Hello World') end. Senior year in College (defun hello (print (cons 'Hello (list 'World)))) New professional #include void main(void) { char *message[] = {"Hello ", "World"}; int i; for(i = 0; i #include class string { private: int size; char *ptr; public: string() : size(0), ptr(new char('\0')) {} string(const string &s) : size(s.size) { ptr = new char[size + 1]; strcpy(ptr, s.ptr); } ~string() { delete [] ptr; } friend ostream &operator ); importheader(); importheader(); importheader("pshlo.h"); importheader("shlo.hxx"); importheader("mycls.hxx"); // needed typelibs
importlib("actimp.tlb");
importlib("actexp.tlb");
importlib("thlo.tlb");
[
uuid(2573F891-CFEE-101A-9A9F-00AA00342820),
aggregatable
]
coclass CHello
{
cotype THello;
};
};
#include "ipfix.hxx"
extern HANDLE hEvent;
class CHello : public CHelloBase
{
public:
IPFIX(CLSID_CHello);
CHello(IUnknown *pUnk);
~CHello();
HRESULT __stdcall PrintSz(LPWSTR pwszString);
private:
static int cObjRef;
};
#include
#include
#include
#include
#include "thlo.h"
#include "pshlo.h"
#include "shlo.hxx"
#include "mycls.hxx"
int CHello::cObjRef = 0;
CHello::CHello(IUnknown *pUnk) : CHelloBase(pUnk)
{
cObjRef++;
return;
}
HRESULT __stdcall CHello::PrintSz(LPWSTR pwszString)
{
printf("%ws\n", pwszString);
return(ResultFromScode(S_OK));
}
CHello::~CHello(void)
{ // when the object count goes to zero, stop the server
cObjRef--;
if( cObjRef == 0 )
PulseEvent(hEvent);
return;
}
#include
#include
#include "pshlo.h"
#include "shlo.hxx"
#include "mycls.hxx"
HANDLE hEvent;
int _cdecl main(
int argc,
char * argv[]
) {
ULONG ulRef;
DWORD dwRegistration;
CHelloCF *pCF = new CHelloCF();
hEvent = CreateEvent(NULL, FALSE, FALSE, NULL); // Initialize the OLE libraries
CoInitializeEx(NULL, COINIT_MULTITHREADED);
CoRegisterClassObject(CLSID_CHello, pCF, CLSCTX_LOCAL_SERVER,
REGCLS_MULTIPLEUSE, &dwRegistration); // wait on an event to stop
WaitForSingleObject(hEvent, INFINITE); // revoke and release the class object
CoRevokeClassObject(dwRegistration);
ulRef = pCF->Release(); // Tell OLE we are going away.
CoUninitialize();
return(0); }
extern CLSID CLSID_CHello;
extern UUID LIBID_CHelloLib;
CLSID CLSID_CHello = { /* 2573F891-CFEE-101A-9A9F-00AA00342820 */
0x2573F891,
0xCFEE,
0x101A,
{ 0x9A, 0x9F, 0x00, 0xAA, 0x00, 0x34, 0x28, 0x20 }
};
UUID LIBID_CHelloLib = { /* 2573F890-CFEE-101A-9A9F-00AA00342820 */
0x2573F890,
0xCFEE,
0x101A,
{ 0x9A, 0x9F, 0x00, 0xAA, 0x00, 0x34, 0x28, 0x20 }
};
#include
#include
#include
#include
#include
#include "pshlo.h"
#include "shlo.hxx"
#include "clsid.h"
int _cdecl main(
int argc,
char * argv[]
) {
HRESULT hRslt;
IHello *pHello;
ULONG ulCnt;
IMoniker * pmk;
WCHAR wcsT[_MAX_PATH];
WCHAR wcsPath[2 * _MAX_PATH]; // get object path
wcsPath[0] = '\0';
wcsT[0] = '\0';
if( argc > 1) {
mbstowcs(wcsPath, argv[1], strlen(argv[1]) + 1);
wcsupr(wcsPath);
}
else {
fprintf(stderr, "Object path must be specified\n");
return(1);
} // get print string
if(argc > 2)
mbstowcs(wcsT, argv[2], strlen(argv[2]) + 1);
else
wcscpy(wcsT, L"Hello World");
printf("Linking to object %ws\n", wcsPath);
printf("Text String %ws\n", wcsT); // Initialize the OLE libraries
hRslt = CoInitializeEx(NULL, COINIT_MULTITHREADED);
if(SUCCEEDED(hRslt)) {
hRslt = CreateFileMoniker(wcsPath, &pmk);
if(SUCCEEDED(hRslt))
hRslt = BindMoniker(pmk, 0, IID_IHello, (void **)&pHello);
if(SUCCEEDED(hRslt)) { // print a string out
pHello->PrintSz(wcsT);
Sleep(2000);
ulCnt = pHello->
The first thing I thought of when I read the title was keyword bookmarklets:t ml
http://www.squarefree.com/bookmarklets/keywords.h
One of those things that seems could be made more useful with some "Ajax" here and there.
Vaguely along these lines, I've had a desire for a while for screen scraping type work. There are small bits of info I want to grab from multiple web pages and combine them into a single page. Further complicating the issue, there may be POSTDATA and cookie files I need to provide, possibly with different versions of a single cookie file.
Is this possible without full-on progamming?
Ooh, a sarcasm detector. Oh, that's a real useful invention.
So does this mean pics come up in asci ?
You are in a dark place. You are likely to be eaten by a spammer.
I am totally failing to see the point of this except for people who want to dare a shave from Occam's Razor and introduce more chance of user input error, waste time getting away from what average users want and need, and just generally engage in more geekery.
Wouldn't Lynx be easier?
If my grammar and spelling are off, I am [distracted/tired/careless] (take your pick)
Yeah, because "Interface Design by Committee" has worked so well in the past. One only needs to look at the W3C to see how the interface design by committee leads to a cluster fuck. Now throw into the committee every person that has an internet connection.
I'll wait for the sequel.
almost as if somebody slapped a CLI onto WWW::Mechanize.
Cut that out, or I will ship you to Norilsk in a box.
Take a look at our good'ol friend Google:
Take a look at operators: http://www.google.com/help/features.html
i've been doing this forever in the form of ridiculously small shell scripts which launch firefox and plug command line args into a url. it works better as i can use the term window that i already have open instead of launching a browser, finding a site, figuring out the command.....
eg weather -h 48237 finds my local weather by the hour
no i have not shot my gun in the air and gone 'Ahh!'
rm -r microsoft.com
(oh please oh please oh please...)
Modern copyright is theft of culture from everyone and it retards the progress of the useful arts and sciences.
I had forgotten all about that idiot Jon Katz until I tried his jonkatz command. Wowzers, that's handy!
Yes, there have been apps that have done this forever. But they've all been standalone, and don't travel with you between machines. I wrote a post about interfacting with YubNub via Monad here: http://www.leeholmes.com/blog/MSHAndYubNubACommuni tyCommandline.aspx
It all goes downhill from first post
For example, try
pop canada
and
gdp canada
It would be so cool if I could type:
"gg:String" and get the google search results on String.
Or:
"ggl:Blah" and go directly to the first match!
Or even add my own commands, all from my very own control center!
It all goes downhill from first post
I was hoping it would be something like this - something I cooked up long ago for Xanga.com.
http://mycroft.mozdev.org/download.html?name=yubnu b&submitform=Find+search+plugins Thought I would point out there is a great firefox plugin for this website.
It should be noted, with all the Ruby/Rails hype lately, that this is indeed a Ruby on Rails app. It was originally submitted as project in the www.railsday.com competetion (to see what can be done with Rails in 24hours).
[Ding dang dong]
[Psssssssssh]
Player 1: I hate smoke grenades
[Pow!]
Player 2: pwnz0rd!
Player 1 (Spectator): Wallhax!
Elsewhere, Kazaa has developed a command-line version of its file download capabilities:
"GET AmericanIdiot.mp3"
[Duh duh-duh DUH duh-duh duh duh-duh duh duh]
"Don't wanna be an American Idiot..."
I have been using Perl for years to do simple command line scripts to extract information from web sites. From foreign exchange rates, to looking up owners/addresses of incoming phone numbers, or abreviated weather forecasts or train schedules.
I began when my old laptop took too long to start a web browser. But even though I now always have a web browser window open, typing "xchange 1432 EUR" is still much faster and simpler than navigating to the appropriate web site and filling in forms.
I'm actually surprised to see such an idea posted here as news. I thought that anybody who knows some scripting has always been doing this sort of thing.
Of course, Perl is ideal for this sort of thing, with regular expressions and all the CPAN modules (LWP::Simple, and dozens of others more specialized).
Yubnub actually defaults to using google, so it's basically just a way to customize google. I figure if yubnub gets any more attention google will come in with the spikey flail to crush some skulls.
I just thought I'd mention that I've been working on a similar site, called QueryAlong. The objective is a little different, but the end result is similar.
QueryAlongUsers can submit search engines and other queriable urls. I think QueryAlong's model is a little more complete too since it can handle multiple named parameters.
For instance if you want to find all the queries that handle Zipcodes as their parameter:
View Zipcode queriesTake a look, sign up and play with it a bit. It's still being developed, and it resides on a server in our closet, but I think it's worth looking into.
Also I'd love some feedback, this is a little project I've been toying with for a while.
This hack converts the Google search field built in to Safari into a YubNub search/"command line" field.
/Applications/Safari.app/Contents/MacOS/Safari
Close Safari. Open this file in a hex editor such as HexEdit, or open it in vi if you know how to use vi:
after creating backup of the file for just-in-case, replace the ASCII string:
http://.google.com/%25@?q=%25@&ie=UTF-8&oe=UTF-8
with this ASCII string:
http://yubnub.org/parser/parse?command=%25@
For this hack I used this hint.
I never noticed that, very cool.
Ironically, the word ironically is often used incorrectly.
Everything and their mom is going "social" these days. Doesn't anyone just want to be left alone any more?
Several web browsers already have this sort of feature built-in. Konqueror and Firefox, for example.
If you are using Konqueror, type the following into the location field and see what comes up:
... now go to yubnub.org, find the search field, type in 'wp slashdot' and click 'submit'.
Like yubnub.org, you can make your own shortcuts. I suppose the idea of yubnub is that you can share these things with friends through their site, but you could just as easily share the shortcut and it would ultimately be easier to use the second time around...
Actually, my firefox bookmarks (and relevant keywords) travel with me just fine thanks to bookmarks synchronizer. Of course, now I just have to figure out where that japanese developer's website has gone so I can find new updates to the code if he posts them. Heaven forbid a new ff release break the current stuff. ;-)
Isn't this just command-line XML-RPC?
Why not simply build a command line tool that leverages what developers have already in place?
I had a look at the site. What is it that is new and/or interesting in this? I went to the "golden eggs" page, which is supposed to list the most interesting commands. What I see there falls into one of 2 categories:
- a complicated way of executing a normal shell command (di ("Domain (WHOIS) Information using coolwhois.com") seems to do exactly the same as a modern whois, ipinfo-url looks like a a lame version of host, etc.),
or
- it does the same as adding a search engine to your Firefox search bar.
Have I missed something?
I'd like a full set of vi commands for the web. :wq (save as bookmark and exit)
ZZ (here comes the boss, kill the porn) :set nu (add line numbers to page) :set list (same as view source)
/
n next search term
p previous search term
AOL Keyword: commandline.
Ring a bell, anyone?
Google buys it?
The days of telnetting into university servers, using ARCHIE to dig through text files, even using 3270 emulation to log into mainframes. I cut my teeth on that stuff (broke a nail too). They quietly faded away 10-12 years ago. When the web came out I was really down on it. How "dare" someone make the internet easier! Gradually I gave in to it and learned to like certain aspects of it too. How is this going to work? The majority of internet users have never even seen the command line on their XP boxes. This is one of the big stumbling blocks to getting Linux distro's that the uninitated will install and use. One of the common complaints is that the command line stuff is "too hard". I work with other engineers who still do not know how to narrow down a search in google for phrases by using the "quotes". Let's call it CLI (command line interface) so it has a snazzy name that will draw people to it.
Tisha Hayes
If you read this blog (http://mradulovich.blogspot.com/2005/03/command-l ine-strikes-back.html), you'll see that the web command line already exists. Why do we need anything else?
This reminds me of Dozomo, that 'internet commandline' created for the 24hour Dotcom. The two basically do the same thing afaik.
or Firefox's Keyword Bookmarks?
I've been doing command-line interaction with simple shell scripts containing one Curl cmmand for ages.
Bad enough there are still people who can't wrap their heads around pointing and clicking - this will be a tech support nightmare if it becomes mainstream. I'm having a vision from the old dos days, when my dad had pages and pages of commands taped all over his study, and I still had to remind him of "cd.." 3 times a week.
Single? Canadian? We can help. Visit http://www.l
For IE6 you add it to the registry: create a new key under
HKCU\Software\Microsoft\Internet Explorer\SearchUrl
and give it the name of the command (i.e. "g" for google).
Set the default string value under the key to the URL to use; %s gets replaced by what you type after the command. For the google example above, set the value to "http://www.google.com/search?q=%s"
You can also do character->string mappings; create a string value whose name is the character to replace, and whose value is a replacement string. Most common are things like ("&", "%26").
Point and click on the web started with Mozilla in 1993. Before that, it was all command line, for the first 20 years or so...wasn't it?
Shell Users' Revolutionary Front Rage Against the Web
, w3m
Once again the *nix world has had this taken care of for years.
I posted about this an AC earlier today, but now that I'm back at home time to give some more info.
From the website:
"Surfraw provides a fast unix command line interface to a variety of
popular WWW search engines and other artifacts of power. It reclaims
google, altavista, babelfish, dejanews, freshmeat, research index,
slashdot and many others from the false-prophet, pox-infested heathen
lands of html-forms, placing these wonders where they belong, deep in
unix heartland, as god loving extensions to the shell.
Surfraw abstracts the browser away from input. Doing so lets it get on
with what it's good at. Browsing. Interpretation of linguistic forms
is handed back to the shell, which is what it, and human beings are
good at. Combined with netscape-remote or incremental text browsers,
such as links (http://artax.karlin.mff.cuni.cz/~mikulas/links/)
(http://www.w3m.org/), and screen(1) a Surfraw liberateur is capable
of navigating speeds that leave GUI tainted idolaters agape with fear
and wonder.
For example:
$ ask why is jeeves gay?
$ google -results=100 RMS, GNU, which is sinner, which is sin?
$ austlii -method=phrase dog like
$ rhyme -method=perfect Julian"
And obviously you can hack this to make it work with whatever website you could want to use. Much better than this YubNub stuff from my glance at it.
Anti-social? My code is just platform-specific.
with a marginally less silly URL?
combine the power of the command line and the web by using wget. :)
C:\Internet\www\>dir ....
999*10^99999 files ([overflow] bytes)
Well, yes ... there have been means to share your own search urls with yourself for quite some time as well. USB Keychains, an FTP account somewhere, rsync, or one of the many dotcoms that busted.
The point is still that it's not social. Until Firefox (and all other browsers, for that matter) can merge them from a public source and promote the good ones, it's still your custom, private, solution.
It all goes downhill from first post
I haven't seen any mention of DQSD yet, so I figured I'd link it. Embeds a searchbar into your Windows taskbar, and comes with a HUGE set of default commands, stored in configuration files. Then you can add your own basic aliases, or write small scripts into XML files. Pretty cool.
Hopefully you're still checking the responses to your comment. After I read your comment yesterday, I started to think about how cool it would be to have a console like that. I couldn't find one, so I scrounged one together. If you want to follow the instructions, I posted them here:
o le-la-quake.html
http://toveness.blogspot.com/2005/06/windows-cons
Hopefully you can find this useful!
No, it isn't social. But it travels. For social ones that travel, how about del.icio.us?
Yubnub's interesting, but I don't see it as groundbreaking.