Slashdot Mirror
John Patrick: ENUM is a Really Big Deal
penciling_in writes "John Patrick, former vice
president of Internet technology at IBM, says 'ENUM is a really big deal'. Here
is what he has to say on CircleID about this: 'Basically, ENUM
is a protocol that will make it possible to converge the Public Switched
Telephone Network (PSTN)
and the Internet. In other words, a telephone number can get you to a Web
service -- telephone number in, URL out. The idea can be extremely useful when
you consider that most telephones are limited to twelve keys on a keypad. Every
try to enter your alphanumeric login ID and password to a web site on a cell
phone or Personal Digital Assistant? It is next to impossible! The biggest
impact of ENUM will probably be for Voice Over IP (VoIP).
In fact, it could be the tipping point.'"
This message brought to you by 1-800-SLASHDOT
Any fool can talk, but it takes a wise man to listen.
Unless you are actually enumerating the values of a user-defined type, it is safer and simply better programming practice to use the const keyword instead of the deprecated enum hack.
And I use VoIP now and it saves me a lot of money in long distance charges. Whether the service provider is using ENUM, I don't know, but I do know that as an end user I don't care one way or the other. Const, Enum, it's all the same to me (as a user).
Next stop IPv6 and having a number for everything...
"Hello this is the refrigerator, the Answering machine is broken, so speak slowly and I will put on a sticky note... with your message"
Food not Bombs is a nice platitude but it breaks down when you notice that the Bombees are usually well fed
And they want to fucking throw phone #'s in the mix, too?
ENUM is good when the PBX is not authoritative for a number. For example, if someone dials a number inside a company, if that number is for another extension in the company you send the call there. It is only when this number goes outside of the company that it is useful.
BTW, It also plays well with SIP (RFC 3261), because that number can map to a SIP URI sip:user@domain to route a call. Neato!
--------
Free your mind.
remember a number like 435364-5464562 for a website or "http://www.slashdot.org"?
when you spend time typing on a cellphones pad it gets much faster and quite frankly words are much easier to remember which is what ultimately matters. the difference between a number and a word or two doesn't really matter(it's just double the button punches at maximum usually, in time it's just 15 secs instead of 10, numbers pulled out of my hat).
oh yeah i've done my fair share of cellphone thumbing, including numerous slashdot posts and thousands of lines of irc. but the freaking teenagers are still faster than me.
world was created 5 seconds before this post as it is.
I don't see how this needs an extra protocol. For instance, you could type in Slashdot's address as 66*35*250*150#.
No reason to make it more complicated than it is.
Besides, old technologies such as POTS (Telephones) and TV are frequently regulated (and taxed!) by governments in a way that makes them unsuitable for open content. The very thought of connecting these very taxable technologies to the Internet scares me.
ENUM is based on trust, in the sense that your ENUM server won't send you off to the phone call version of goatse. So far we have other systems based on trust like the DNS system (which is being systematically undermined by those designated to care for it, but I digress). It is only through trust that we are able to hit the enter key without fear that our DNS server won't send us to tubgirl or comp-u-geek, and so far it's worked out quite well.
However, I don't trust people who have two first names and no last name, "John Patrick".
The whole idea behind having to dial a *number* to get in touch with someone is so antiquated that it's hard to believe that in this year 2003 we are still using it as our primary means of communication. It's like a giant cow just standing there eating and sleeping and sucking up valuable resources and farting toxic methane gas which is warming our atmosphere making this world unlivable.
If we can get VoIP, we can finally tip over this flatulent bovine and bring ourselves into the 21st century. I shouldn't have to dial a number to connect to someone. No one ought to be reduced to a simple number! I should be able to call an acquaintance using their name or ideally some truly identifying tag like a URL. ENUM is essentially the only system that would allow this kind of connection.
The phone companies are sleeping on their laurels now. It's time to bring the next tech to bear and knock these soon-to-be hamburgers over.
your tax dollars hard at work? i am never voting for more spending in technology again so kids can get high on cough syrup and post this shit.
Cirruz
Humanitarians eat humanitables.;)
Is this a sigs-optional kind of place? 'Cause I am totally down with that if you know what I mean.
Drop the current three-letters-per-digit phone scheme and add six new numbers to every phone. Let's call them A, B, C, D, E, and F. Let's face it, for 6 extra digits, we get 2.5 times the phone numbers. Add 2 extra digits to the current limit of 10 digits per phone number, and it will be IPv6 compatible. And let's face it, there no more dificulty in remembering phone numbers if they include Alpha characters.
"The whole idea behind having to dial a *number* to get in touch with someone is so antiquated that it's hard to believe that in this year 2003 we are still using it as our primary means of communication."
It's not been my "primary method" for about 5 years - I always dial stored numbers, looked up by name (apart from the first time I call someone, and assuming they didn't send my phone a business card).
"It's like a giant cow just standing there eating and sleeping and sucking up valuable resources and farting toxic methane"
It's almost exactly like that, yes.
Not sure how this could be made transparently. first thought is, it's a new DNS system, but besides changing all our DNS software, we'd also need to make applications aware of it... that's a lot of code. Maybe we'll just make the change for web browsers, being that the web-surfing non-tech public are the ones who'll enjoy it most.
If the RFC is correct and it's true that TCP/IP can be implemented using carrier pigeons, how would one go about making the pigeon to wire interface? And if it is possible, how do you handle implementing ENUM on top of that?
Can the pigeon poop be recycled somehow?
mmmmmm, phone fumes. . .
-nando
"No one ought to be reduced to a simple number!" what's your Social Security Number?
Honey - it's ten o'clock,
Junior should be home by now.
Dear - so ring his cell phone and
tell him to come home!
Never mind, I just ran traceroute on his
phone... He is... uh-oh..
Sisy - Oh, he is so grounded... hehe...
Telephone numbers are almost as long as numeric IP addresses. Why not get rid of DNS and key in the addresses? And issue paper "Internet books" where we could look up the IP addresses of sites we wanted to access?
I'm not sure what aspect of cognitive psychology explains this, but I and many of my middle-aged friends acknowledge having problems remembering local phone numbers now that in our state we are required to dial area codes on all calls.
I'm talking about short-term memory here. Forget long-term memory, I haven't stored telephone numbers in my brain in years; I have a Palm for that.
What I'm saying is that I can look in a telephone book, see the local number 762-####, and even though all local numbers have the area code 781, remembering to dial 781 (and dialing it) put enough extra strain on my enfeebled brain that by the time I've dialed the 781 and the 762, I can't remember the last four digits any more.
Actually I dial 1-781-762 because, for some unknown reason, the power that be have decided that a) you must dial a 1 if the call is a local toll call; b) there is no simple algorithm for determining whether or not an exchange in the 781 area code is a local toll call or not; c) if I don't dial the 1 some of my calls will not go through; d) if I always dial 1 all of them go through; e) dialling the 1 does not change the cost of the call, so... well, you get the idea.
Information theory says that the "1" carries no information, and the "781"--being the same on 95% of my calls--carries less than a single bit of information. That "1781" should only amount to a fraction of a digit's worth of memory strain.
Well, tell it to my brain, because it apparently doesn't know how to do data compression on the leading digits of telephone numbers.
Even if I keep the telephone book in my hand, by the time I've finished dialing the 1-781-762, my eyes have moved off the page and it takes me a good two or three seconds to reacquire the listing visually, scan carefully horizontally to make sure I'm picking up the number that's on the same line as the name, get those last four digits, and transfer my attention back to the the dial.
The phone system seems to be getting less and less patient even as telephone numbers get longer and longer. Even though the equipment must cost orders of magnitude less than it did twenty years ago, the engineers that set the timeouts apparently can't stand having their precious equipment tied up. The result is that sometimes the phone times out and abort the call before I can key in the last four digits.
Yes, I considered posting this as an Anonymous Coward. No, I don't have any memory-related mental diseases--and I'm not all that old.
But the prospect of tying websites to telephone numbers strikes me as the dumbest idea I've heard in a long time.
"How to Do Nothing," kids activities, back in print!
"Jenny, Jenny, who can I turn to? 9.0.3.5.7.6.8.e164.arpa" doesn't have the same ring to it.
One CPU cycle wasted on digital restrictions management is ONE TOO MANY.
Cool, maybe now I can get a dns style address for my phone number that won't have to change when I move houses, or change countries.
Every try to enter your alphanumeric login ID and password to a web site on a cell phone or Personal Digital Assistant? It is next to impossible!
Effective next monday, all Internet users will have their usernames and passwords converted to 32-digit numbers to comply with telephone-Internet convergence. Use them to login to our website at http://2529850985513857918375981751. Your username and password are:
Username: 72946835 56198569 01854984 91856914
Password: 57105710 19158294 19469819 14691749
This is done for your convenience. Have a nice day!
Soon when you dial a wrong number you'll get Verisign on the other end asking you for your credit card number.
It will no doubt be a big deal for VoIP. Why? Because it means that all you have to do in order to associate your IP address with a phone number is to register your phone number with DNS. That's certainly a nice-to-have.
But when they suggest that we could use this to somehow ease the pain of using a PDA/number-pad as a UI to the Internet (read: WWW), and therefore bring about the holy grail of Convergence, that's when my bozometer redlines. OK, so that phone number translates to a DNS name, which in turn resolves to an IP address which, say, takes me to a website. Now what? I still have to login!
Maybe the plan was that I'd make up a special phone number for myself that would map to a special IP address that would redirect me to an URL that would post my login information for me. But why would I ever want to do that? Have they completely forgotten that logins are a security measure, not just an inconvenience to be circumvented via shortcuts?
Don't get me wrong; it's nice that we can make it easier to enter canonical URLs using a numeric keypad. I appreciate that. But let's not get too excited about this. As a UI enhancement, ENUM a trivial, evolutionary step forward. It isn't, and won't ever be, a "really big deal."
-Dan
When I moderate, I only use "-1, Overrated". That way, I never get meta-moderated!
seems like an attempt to wrestle control away from "the Internet" and back to the telcos. Sure it has some uses, and therefore needs IETF standards eventually, but this hype about it being the next big thing? Sounds like an attempt to re-establish phone numbers as the primary business identifier, and return control of small business to the regulated monopolies and their lawyers. Reminds me of the current "Yellow Page" advertising system.... $700 per month for an ad, and it can't contain a website address larger than 12 point type. Dinosaurs are still might beasts, right up until they die.
Of course, it would be every so convenient if they mapped unused numbers to a brief message and an advertising opportunity. Verisign could help.
Hmm. So rather than typing 410-455-xxxx to speak to me, you could type "Tom Swiss, Baltimore, Maryland, USA" on your terminal. Then wait for a lookup. Then have to select "Which Tom Swiss? Thomas M. Swiss in Catonsville, or Thomas C. Swiss in Towson?" Doesn't sound all that better than having a unique numeric ID.
Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood
Either that, or there would need to be strong rules to give people control over the domains to which "their" numbers point -- or at least rules ensuring that they point (for SIP / Internet Telephony purposes) to the right line. I suppose personalised web / mail servers at those domains could be value-added services from whoever owns the domains to which the telephone numbers eventually resolve.
Of course if they mess it up too badly, the rest of us (or the ITU) can use some other domain for "enum" (assuming it's a good idea).
Paul "Say no to feeping creaturism"
There are many issues to be worked out
Technically, the QoS (latency) over the low quality lines, seems something of an issue. I read somewhere that anything more than 0.1 sec latency makes conversation noticeably less spontaneous. Maybe it will encourage us all to listen to others without interrupting them:)
Politically, the regulated monopolies that provide local telephone service will be difficult to interface with. They'll demand money for access to the offices (some people have stories about this), but are also obligated to maintain lines and obey a slew of regulations, such as providing service to backwater areas, emergency service, tacking on a whole zoo of fees and taxes.
"Provided by the management for your protection."
1. bango.net - doing this sort of thing for more years than I care to remember; initially as a dot.com "turn your phone number into a web link!" idea that no-one bought into, now as an "instead of remembering your WAP site address, users can simply type in a string of unmemorable digits to get at your phone pr0n site" sort of thing. (Anyone who's got "Mr Badger's mobile porno" as a customer must be doing well)
2. Being able to turn numbers into resources locators has nothing to do with typing in users and passwords on mobiles, unless I've missed something.
The point with easily memorable phone numbers is usually that they sound musical or rhyme, or that they make a logical pattern on the keypad of a typical telephone. When you can have words instead of numbers, it strikes me that web addresses are already far more easily memorable than the typical telephone number which is just a random string of digits, maybe with a few double digits, or if you're lucky, triple digits.
The tipping point for VoIP will be going to the local discount store and being able to buy a standards-compliant phone with ethernet and/or POTS (depending on the desire to bridge to POTS) for less than $50. Add slightly more money for greater features (more POTS lines, 802.11, cordless, voicemail, etc).
In other words, the standards for VoIP need to crystalize to the point where interoperability with gatekeepers and other switch-like devices is a given, and not some game of vendor lock-in. I should be able to buy a Samsung/Apex/Sampo/$cheap_asian_brand VoIP phone, plug it into my ethernet network and have it just plain work with other VoIP phones, bridges, etc.
How or what the various numbering protocols work isn't the tipping point. The PSTN is too big and complex and the legacy devices too numerous to think that a new, acronym-loaded, buzzword-compliant numbering scheme will make a difference. For VoIP to matter it must initially be transparent to the POTS world, and that means telephone numbers and bridging.
Vonage is on the right track here as a bridging service. Their POTS bridge device is on the right track at least conceptually, although I can't comment on its protocol neutrality.
Is a telephony invention that permits access to my voicemail (enter the PIN code) using a hands-free headset while on the road with no button pushing on the phone.
Remember - this controls what happens when people attempt to reach *your" phone number!
Anything else, like other IP based services that could be offered on different ports from a server with the same URL is icing on the cake (or a huge revenue opportunity / regulatory nightmare / security problem / take your pick).
Paul "Say no to feeping creaturism"
I recall ENUM was patent encumbered, which is why it had not received wide use previously. Is this still the case?
EMUN actually sounds transitional to me. It's a bridge technology to integrate the phohe system, and then at some point, when we have deployed better interaction models for telephony, we'll go to a broader, more flexible URI scheme.
Actually, I also think it's not a complete integration. There have to be other things you can integrate with a live stream of data. Digital signatures would be one. Perhaps some open (and better designed) analog of Powerpoint might be another.
'In knowledge is power, in wisdom humility.'
It's not, yet, about accessing websites as such from a phone, or providing a harder to remember alternative to URLs.
Though with sophisticated phones (approaching PCs), that distinction could begin to blur...
Paul "Say no to feeping creaturism"
O'WONDERWe're working on it.
This message would have been brought to you sooner by 1-900-SLASHDOT ( 0891 SLASHDOT in the UK... )
Paul "Say no to feeping creaturism"
Dear 265275*,
Thank you for your input in comment 7259667. Please keep up the valuable commentary.
Kindest Regards,
514750
* Please note that when IPv6 becomes more widely released, a new value will be assigned to uniquely identify you.
Sure I'm paranoid, but am I paranoid enough?
Well, we're lucky they chose the .arpa toplevel, otherwise each wrong phone number would have been answered by Verisign's sitefinder ...
What do I get when I accidently dial goatse?
Man, I feel stupid.
/. blurb, and skimmed the related article, all the while nodding to myself and thinking "Hell yeah...I use ENUM as the data type in MySQL all the time, and I've been doing it for years. I am so ahead of the curve. I rock."
I read over the
D'oh. Wrong ENUM.
-Waldo Jaquith
This seems to be another pump and dump technology. Calling a web service based on the phone number? What about the inputs to the web service? what about authentication? What about security? This seems to be an incomplete solution to a problem that dosent exist.
Ok IP v4 is allready out of address space and Phonenumber space is less than the adress space of IPv4... Should I go on?
:) After all IPv6 should be here soon and it has About 340,282,366,920,938,463,463,374,607,431,768,211,45 6 addresses...
:) I would have to remember over 3 times the digit to her phone number because as usual i will undoubtly loose the piece of paper she worte it on or spill a drink on it or some other disaster.
:)
:)
I have no clue why someone would think that using a 68 digit phone number to get to a web url that is only 14 easy to remember chars long would be a good thing
For anything new to be accepted and widely used... It needs to be atleast as convient as what is currently inplace... ipv6 is "just around the corner" so I think haveing to remember a 39 digit phone number doesn't meet.
Also if on a lucky night at the bar i find a woman drink enough to give me her phone number
But what would be a ground breaking idea is.. PNS... linda like DNS but for phone numbers instead of domain names... But then verisign would redirect me if that phone number wasn't registered to someone lonely enough to pay them money to have people call them cause they think thats what I want
ps
I know they have something similiar to DNS for phones.. its called Directory assistance
Who needs WiFi when we can have Packet Over Sheep! http://datacomm.org/PoS-InternetDraft.txt
Oh yeah. What about "bookmarking" your frequently used numbers kind of like you bookmark your websites.
This isn't a replacement or even an add-on. It uses stuff we already have to offer more to the internet. For all those who don't see the point and think that using numbers is a step backwards, think about this: the "old-fashioned" method of using numbers has worked well for years with the telephone. people already have the habit to use phone numbers. those habits can now be used for the internet. And here's the thing I like: all companies have to do is register their existing phone numbers in this and instantly everyone in the country has an intenet directory in thier house. It's called a phone book. and there's even one that sorts by category! :)
Also, everything that now stores phone numbers can be used for internet too (address books, bussiness cards, etc.) Not to mention you no longer need a full keybord to use a web browser (think web-tv's, net appliances, etc.)
AB HOC POSSUM VIDERE DOMUM TUUM
Geez - whoever modded as funny, has 1) no real sense of humor, 2) is ignorant on technology.
You mean kind of like speed dial? And to think, we had this technology years before Al Gore even invented the web.
Don't blame me; I'm never given mod points.
If you want a number, use decimal IP notation.
If you want a name, fix the bloody phone software to make it easier to enter URLs. Maybe even make it suggest completions from the phone companies' DNS cache rather than the phones' internal dictionary.
What's the betting IBM have a patent on this ENUM bollocks?
Idiot. This sort of tripe makes me want to gobble the innards of a cows stomach for pleasure.