Slashdot Mirror
75% of Network Connections Not From Browsers
Trailwalker writes "An article at BigBlueBall.com states that 75% of web connections do not use a browser. IM and P2P applications are used instead." While surprising, this is probably more indicative of how instant messaging has been able to complement and/or replace email in recent times.
IM has less spam than e-mail!
How often do you get pr0n/anime/music/movies over http? That's what I thought.
My life's goal is to get a score of +3!
One would think that spam and email worms constitutes a significant portion of that 75%.
The World is Yours.
What about gaming? Many connections from that :)
This space is not for rent.
I would think that msblast makes up a larger amount of the applications with network connections. I work for an ISP and there a still many customers who are afflicted with that virus.
Well,
My 4000 song lawrence welk collection just got found out....
I thought I was doing the music world a favor, renaming song , and artists like shaggy with welk tunes behind em...
Guess Ill have to go back to spam for a living....
I don't find this surprising at all... think about it... most people probably have 1-2 AIM windows open for each browser window, and sometimes have an internet radio or something else running at the same time.
Serously, the second AIM puts ICQ-like away messaging in, email is dead.
It's much more convenient to just start up a program, punch a few numbers, and see who is on for real-time conversations and get all the messeges they sent while you were away. Further, with the increased use of Webmail based email servers, it is becoming more and more inconvenient to check it... It's become rare to have a non-isp provided email account be POP3 by default.
Is AIM ready for the spotlight? Not quite... AIM, the popular one with teens, i'd say, still needs to work on privacy, logging, and message ability while someone isn't logged on. ICQ is a more perfected clinet, but the settings and UI are much too complicated for a novice.
Until recently, I only checked my email every couple weeks.. i'm on AIM almost constantly.
The article states that "76 percent of active Web surfers access the Internet using a non-browser based Internet application." I take this to mean not that only 24 percent of traffic is HTTP traffic but that 76 percent of people who use the Web use something else as well.
Seriously - how do they come up with this number? Are they packet sniffing the entire internet?
I'd like to know their method before I would worry about their conculsion.
Weaselmancer
Weaselmancer
rediculous.
After 'bigblueball.com' gets /.'ed I think their 75% claim will change a little.
A clever person solves a problem. A wise person avoids it. -- Einstein
Does it even matter anymore?
The article states that 75% of users use non-browser applications to access the internet.
It DOES NOT say that 75% of the connections are made by non-browser applications.
There is a difference, and I blame the lack of any sober editors at Slashdot today for this getting through.
how do they come up with this number?
From the article: "Source: Nielsen//NetRatings, December 2003". More information on Nielsen's products may interest you.
If Nielsen's net ratings work anything like their TV ratings, then lucky families get paid to put a spybox between the cable modem and the home router, with full knowledge and consent of what's going on. I'd expect an airtight privacy policy; Nielsen has provided TV ratings for over a decade.
Putting statistics in terms of connections from unique users doesn't quickly mean the importance of these applications, and also usage patterns of internet users in general.
Also, Windows Media Player and Realplayer establish browser connections to their media homepages anyway; does it count as a browser connection? In most cases, it is not even initiated by the user (the user might be wanting to play MP3's).
What about e-mail? It is very important and widely used by everyone; but it doesn't even make the list!
thinking of a school.... they once did a network traffic tests and it was something like 76% was aim/icq (though i doubt icq) yim or msn. 4% online games (program games not web based). and around 20% web browsing. no were near surprising to me.
"Internet users spent an average of three hours and 37 minutes per month using Internet applications."
I don't mean to badger at statistics without seeing their complete methodology, but many people just leave their IM client connected. And don't WMP and RealPlayer phone home? So it seems hard to hold to numbers without specifying an activity which can more easily be linked with HCI time.
The truth is that IM and music players probably are the big draw for most people, which is the conclusion in article title ("Instant Messaging and Media Players are Primary Internet Applications"). Of course, hasn't the Internet always been a majority of activity not directly related to a web browser? Is this news, or a new (more realistic) perspective?
After reading the article, I noted that Windows Media Player is the #1 application accessing the internet. All I can ask is, why?
Why does a media player need to connect to a server so frequently? What information is it sending out? What good does Windows Media Player provide the end user, that it is taking up 34.43% of the web connections?
Make me your friend; my fans get +1 comment scores.
Dude, you plagiarized my stuff.
0 27 381
http://slashdot.org/comments.pl?sid=79523&cid=7
How...weird.
Sure, maybe the majority of network traffic may come from other sources than the web, but this doesn't mean that the web plays any less of a part in most people's regular internet activities. On average, I can assume that most P2P connections occur somewhat dynamically (unstable, temporary, etc). For example, it is not uncommon for a single bittorrent download to require 20+ connections to other users. Since the connections of most home users are somewhat unreliable, this many is necessary. So of course this means more P2P traffic than that of http, but we have to keep in mind that what could have been retrieved with multiple P2P connections would only take one steady connection with a web server. And if IM traffic is measured by the connections between client and central server, well, that ought to be pretty many as well.
For sure. Not to mention, a good portion of that 25% is someone actually viewing the data that's being transferred - I'd say the bulk of the other 75% is binary data, e-mail spam, etc.
- It's not the Macs I hate. It's Digg users. -
Is that unambiguous enough?
"I assumed blithely that there were no elves out there in the darkness"
...for the common man like STunnel, FreeSWAN, or OpenVPN, how long can it be before people are just using private networks between family and friends at home to do IM, P2P or even Windows File Sharing? I've moved in this direction already with my family and friends. All it took was a little of my time to set up SSH clients with Local and Remote forwards that my family and friends initiate connection to my server with. Then they just access the Jabber server I run or, the internal mail server using IMAP, or the recipe database I've created, etc... Since some of my friends and family are Windows bound, I've been able to get them to use the Exodus client for Jabber with cygwin SSH to communicate with me. We even share RDP and VNC sessions. So... what does this have to do with the article? I would argue that there are a good number of people out there doing more than just IM, P2P or web browsing and they are probably doing it via tunneling. It can't be long before this becomes a part of the OS (even for Windows) to allow people to share data in new and very secure/private ways. It's done wonders for the support I offer my friends and family too...
Un-news
These statistics may be derived by bandwidth alone in which i'd probably give it some credibility as video and audio streaming can be extremely bandwidth intensive (especially the former).
One 3 minute streamed video clip at 300 kbps may require as much bandwidth as 100 webpages.
Hmmm... Pie...
How often does the average user's email client (read: Outlook Express/Outlook) check for new messages on the server?
In the same light, how often does IM clients check for new messages? Does it reuse the same connection, or initiate a new connection with the server? (I'm assuming its client->server, due to NATs, firewalls, etc).
Just because the client software is constantly requesting data does not make the software more popular. Statistics about *active* use (say, page hits, email messages, etc) would be more informative.
I wonder if this is because IM and P2P applications are left active all the time. In many cases running whenever the users computer is running.
http://www.kubuntu.org/
The grass is only greener, if you don't take care of your own lawn.
Is there anyone out there that doesn't use at least one of; email, ftp, irc, IM, streaming media, news or ssh outside of a browser at some time...
After performing an in-depth study of the article summary, I estimate that 24% of people don't do any of these things.
Expert in software patents or patent law? Contribute to the ESP wiki!
Thats funny, I was just posting my project (WebP2P) to freshmeat.
The idea is to create a P2P network that actually runs using PHP pages as the peers... technically it would be "pure browser" since tha pages use http to communicate.
meh
Correction to the article headline and link text - The pr from Nielsen doesn't say what percent of connections are via non-web browser software: "Nielsen//NetRatings, reports that three out of every four home and work Internet users, or 76 percent of active Web surfers, access the Internet using a non-browser based Internet application." That doesn't mean that these same users don't use a web browser for the majority of their http connections, rather it says that 76% of active web surfers *ALSO* use Internet applications other than (as in in addition to) web browsers.
My apologies. That article sucks. In fact, it's damn near fraudulent.
"I assumed blithely that there were no elves out there in the darkness"
This article is ridiculously confusing. First of all, it looks like they're saying that 76% of people who use the web also use other programs besides web browsers to get internet content. But it doesn't help that apparently "Internet" has a very particular meaning, which excludes email (otherwise email traffic would have figured in here somewhere) but also miraculously excludes the web, since "Internet Applications" are only used by 76% of web surfers according to this article. This means either that web browsers are not, in fact, "internet applications," or that 24% of web surfers surveyed have direct telepathic connections to the 'net.
Unfortunately, there's no better info available... The company's original press release is pretty much reproduced in its entirety, word for word, by bigblueball news. I hope that Nielsen's clients aren't actually paying for info like this.
Don't you wish your girlfriend was a geek like me?
I also work for an ISP. I too am in charge of an Allot NetEnforcer. I know exactly what you're talking about: peer to peer traffic is accounts for about a 40-45% of our network traffic, which makes it the #1 use of our resources. This makes sense because, honestly, does one really need a broadband connection to send e-mail or load a webpage? I'm not surprised that your customers haven't called. The fact is, they probably won't, unless you disable peer-to-peer traffic completely, because I think most people attribute slow connection speeds on the other user (which is probably the case 90% of the time) rather than the ISP (provided their other types of connections go quickly, of course.)
Something that I suggest you try is what we do: throttle p2p during peak hours (during the day usually) then leave it unrestricted during times of typically low activity (night). As both a devoute kazaa-lite K++ user and an ISP employee, I feel it's a good compromise.
As for your last statement, I tend to disagree. P2P software isn't going away. It's like trying to stop a mouse by plugging up the mouse hole. Things like the NetEnforcer aren't as dumb as a TCP port blocker, but they still have their flaws. The NetEnforcer is able to determine what service is running by examining the TOS bits of the IP headers of traffic. Usually, like port numbers, these don't change. However, I've always wondered what would happen if someone were to create a program that randomly cycled through the possible TOS combinations. My guess is that the connection would still work, but devices/applications like that NetEnforcer would be left clueless. And even if that approach didn't work, things like SSH port tunneling would. Much like the RIAA-IP-blocking versions of Kazaa, I think we can bet that the P2P software designers are going to react rather than roll over and die if what you're saying comes true.
-Grym
Internet users spent an average of three hours and 37 minutes per month using Internet applications. A page like a slashdot article containing some comments takes me more time to read (no internet activity in between) than to download. So, even if they could sniff all our data usage, it would not reflect our real time usage because of differing idle times for each application.
The other 75% of the people are telnetting to port 80 and entering the GET and POST commands by hand.
;)
Well, maybe not ALL of those people are doing that, but the cool ones are
Irritable, left-wing and possibly humorous bumper stickers and t-shirts
The answer is, get better friends. The whole point of IM is you can see whose online and available. For things that are time sensitive it's great. For example, I could email want to go to the movies at 7 tonight? around to all my friends. They could then reply when they get it, which might be the next day, or in 15 minutes. By the time they respond to my email, I might not be in a position I can check mine. With IM, you looks whose on, and talk with them. Furthermore, you can just strike up a chat, if you want to talk to that person, but anyone that gets annoyed if you dont chat with them is just retarded. If you just tell them you are busy at the moment, most people understand. If you dont want to be disturbed unless it's very important, put up an away message. If someone is really annoying, just block them. Most IM clients just show you as logged off, so if you need to talk to that person later, just unblock them. IM is best when used in situations where you would normally have to make a phone call, but don't want to interrupt the person the way a phone does. If IM isn't for you, dont use it. But most people like it the way it is, so I doubt it needs an overhaul. From the sounds of it, you and the people you talk with are trying to use it like a replacement for phone or email, instead of using it for what it's good at.
The article is confusing because it does not define "active user reach". It's easier to understand in this 3 year old similar Neilsen study [PDF}. The table there makes it clear that "active user reach" refers to what percentage of the total population being studied (the "active users") are using the various applications.
The quote from the Nielsen analyst in the current article makes it clear that "active users" are Web surfers, which by definition are people who use browsers.
So the article says that 76% of the web surfers studied also use some other Internet applications, 34.43% of them use Windows Media Player, 20.27% of them use AIM, etc.
Note that this says nothing about what percent of the traffic any of that represents. It seems obvious to me that they cannot be counting email as an "Internet Application" for their survey.
Being Neilsen, they are only interested in applications that can serve advertising. "Reach" means what percentage of web surfers can be reached via advertising delivered through Windows Media Player, AIM, RealPlayer, etc.
It's saying 75% of _Internet_ traffic is not HTTP, as far as I can tell. Back in the old days most Internet traffic was SMTP, NNTP, and FTP. I'm not too surprised if SMTP still is a large part of Internet traffic.
The summary makes it sound like 75% of port 80 connections aren't from web browsers, which would be weird IMHO. (On the other hand, there are lots and lots of web crawlers/spiders out there, so I could believe this stat too, at least for an "average" site.)
The authors of the article clearly do not know the difference between the Internet which are all using as soon as we connect to our ISPs and the World Wide Web which is a linked set of pages of information.
How this simple fact could be missed detracts from, and also complicates, the information contained in the article. It actually makes it meaningless. But then obfuscation is often the reason for press releases!
I highly doubt that 75% of all web connections do not involve browsers. I know that curl and wget are neato, but 75% of all web connections? Nonsense.
On the other hand, saying that 75% of internet connections are not by browsers isn't news at all.
Just pointing out that "web" implies www which implies http or https and nothing else.
Don't become a regular here, you will become retarded. -- Yoda the Retard
Referenced Nielsen article: 76% of browser users have used a streaming data client or an instant messaging client. (dull)
Hmm.
mt