dslreports logo
site
 
    All Forums Hot Topics Gallery
spc

spacer




how-to block ads


Search Topic:
uniqs
16
share rss forum feed


Ben
Premium
join:2007-06-17
Glen Carbon, IL
reply to jcremin

Re: They're not asking the right question.

said by jcremin:

I did a really quick search but didn't find the study I was thinking of. It was from probably 6 months ago or so. I did find this page, which didn't seem to be as in depth of a study (I just skimmed it quick and didn't read the whole thing) but appeared to have a similar conclusion: »gigaom.com/2005/12/20/need-for-speed/

     Thank you for sharing that link.  The information in that link is quite old.  YouTube was almost brand new, and Netflix wasn't streaming video back then as far as I know.

     Although, some of the concepts posed in it are still relevant today.  Comparing the Internet to the road system (a wonderful analogy that helps explain many Internet-related things), I suppose it's like saying that faster speed limits won't mean that people would drive more miles.

     Another thing mentioned in the article is still true today as far as I know.  That many web servers limit the throughput to each individual user.  I got my information from a forum for forum admins that I used to frequent, though it wouldn't surprise me if other web administrators do similar things.

said by jcremin:

Anyway, Yes, I'm sure you can tell a difference between 1.5 megs and 16 if you do video streaming or are are a power user. And if I remember right, the test wasn't focused so much on streaming video as it was on average surfing. But either way, I do you think you would notice much of a difference between 1.5 and 3 or 5 megs? Yes, I would think so.... Now how about between 5 megs and 16? I'm guessing most of the easily identifiable differences quickly disappear.

     Good point.  Basically, you're referring to the concept of diminishing returns.  I used to have a 10Mbps download speed.  Later, it was upgraded to 16Mbps for no extra charge to me.  While that extra speed does come in handy, the difference isn't as significant as when I switched from 1.5Mbps to 10.

     I actually posed this very question to some people, on this site.  It was along the lines of "What do you ever need 20Mbps for?  Isn't 16Mbps enough, since many web sites throttle the speed anyway?"  The answers I got were basically along the lines of how the faster speeds can help in larger households, such as a family of four.  It wasn't something I considered, since I live alone.

said by jcremin:

Either way, my point is simply that the story cries "the sky is falling" because 1) we can't play the speed game on the same level as other countries, most of which are either very small or have insane population densities, and 2) we don't have 50 megs available today, even though nobody really needs it, and it is still 4 years from 2015.

     For the most part, I've agreed with what you said so far, except that I think 3Mbps or 5Mbps is a more appropriate value than 1.5Mbps.  However, anytime someone uses the population density argument, I always have to disagree with them.  If it was population density, then nearly every metro area would have broadband speeds that rival those numbers, since some of these places have densities that are similar to some of these countries.  There's more to it than population density.

jcremin

join:2009-12-22
Siren, WI
kudos:2

said by Ben:

Thank you for sharing that link.  The information in that link is quite old.

Sorry didn't look at the date, but as you mentioned, some of the concepts still apply.

said by Ben:

For the most part, I've agreed with what you said so far, except that I think 3Mbps or 5Mbps is a more appropriate value than 1.5Mbps. 

Yeah, more speed is always nice, but the US also has a lot of rural areas, and I can tell you from running my own rural ISP that it is very hard to push that much speed to everyone. You can't afford put a tower or DSLAM every mile for a dozen homes, so you have to rely on making technology focus on distance rather than speeds. In my area, it is quite expensive to deliver 1.5 megs throughout the area, and 5+ megs is impossible until technologies become affordable enough to deploy more often or can push the speeds farther from the hubs/towers/etc.

said by Ben:

However, anytime someone uses the population density argument, I always have to disagree with them.  If it was population density, then nearly every metro area would have broadband speeds that rival those numbers, since some of these places have densities that are similar to some of these countries.  There's more to it than population density.

That may be somewhat true, but many providers that serve metro areas also serve more rural areas, so you have to take the overall density of the entire area they serve. Rather than charging really low prices in the city and higher prices in the rural area, they quite often charge one flat rate for their tiers so they can make national advertising and promotional materials easier. Because of this, the people in the city pay more than they would otherwise, but it causes the people in rural areas to pay a bit less.

In the same way that many cell companies lose huge buckets of money on rural towers considering how little use they get, but they count on the money they make in the metro areas to make up for it because those people sometimes travel through rural areas and they want a carrier that can cover huge areas.

But yes, there are a LOT more factors than just population density. Someone else (maybe in a different thread) points out that many countries heavily subsidize their infrastructure. Labor rates play a part. The list goes on and on....

pandora
Premium
join:2001-06-01
Outland
kudos:2
Reviews:
·ooma
·Google Voice
·Comcast
·Future Nine Corp..

1 edit

1 recommendation

said by jcremin:

That may be somewhat true, but many providers that serve metro areas also serve more rural areas, so you have to take the overall density of the entire area they serve. Rather than charging really low prices in the city and higher prices in the rural area, they quite often charge one flat rate for their tiers so they can make national advertising and promotional materials easier. B

Imagine what would happen if Comcast, Verizon or AT&T charged less to people in high population density areas vs low population density areas.

The FCC, state and local regulators wouldn't allow it, citizens would be screaming, and Karl would be leading the band marching against it.

This is a no win situation. The U.S. counts internet access differently than some parts of Europe, some countries only count people as not having internet IF they CAN have internet but choose not to. These would be houses passed by a U.S. ISP, it wouldn't measure actual broadband penetration or access by ALL citizens of the OECD nation. This means those that have no access to internet aren't counted at all in some countries outside the U.S. The result makes internet penetration look a lot better than it is. Karl doesn't like to look into those shenanigans.

In the U.S. we have many low population density counties, and serving them is costly. Large utilities who are forced to serve them must cover their costs from the higher profit urban and high density suburban areas.

Our government generally at the federal, state and local levels likes to enforce a duopoly between cable and iLEC for the last mile to customer homes. That enforced duopoly precludes competition. I don't see the FCC, states or local governments changing their belief in a duopoly anytime soon. If AT&T knew back when cable was starting that it would compete directly at all levels of service with them one day, it is doubtful we'd even have the duopoly we do today.

There is no magic in other countries, they have different cultures, and count things differently than we do. The very first thing that needs to be done is to compare apples to apples, all users in each country should be counted and considered as having or not having internet. Actual speed should be considered, not rated speed. I regularly pull down 4x my rated speed on speed tests.

I read a post by nitzan See Profile once where he indicated great speeds in Japan, for JAPAN only traffic. Once he started to point his browser to the U.S. speeds became very slow. Gee, great speeds in country, or in region, but very slow out of the nation? Sounds more like a fast LAN with slow WAN, but Karl will ring that up every time as far superior service than we get in the U.S.

Karl read nitzan See Profile's post here - »Re: US broadband service is costly and constipated. 100 MB/sec in Japan, even 1,000 MB/sec, but once he goes out of country, 2 MB/sec. Wow, I am so jealous (NOT).

Next caps should be considered. Is it really better to have "up to 25 MB/sec down" in France on DSL if you are in Paris and next to the CO but are capped at 5 GB, or if you live 8 blocks away and get 2 MB/sec down, does that still count if it's on the "up to 25 MB/sec down" plan?

Karl, you love to write about this stuff, but please look deeper into the basis for many of the stats you love to cite, sometimes I think you enjoy regurgitating government broadband propaganda (from every country but the U.S.). Why does the rest of the world get a free pass? How about give the same scrutiny to stats from every country that you give to the U.S.? If you were equally skeptical about all the stats you cite, I think the U.S. wouldn't look quite so bad. Though that is just a guess.
--
"People demand freedom of speech as a compensation for the freedom of thought which they seldom use."