I always get a bit nervous about the technical accuracy of an article like this when stuff like the following gets past both the author and editors:

   Fast Ethernet uses the same CDMA/CD as Ethernet

That should be CSMA/CD. They had it right earlier and later on, not sure how this reference got by them, Ars is usually pretty good about nailing this stuff. The article itself is actually pretty darn good.

Also,

   On the other hand, in 1975 few people would have guessed 
   that today's students would go to class carrying 
   affordable computers with 10Gbps ports.

I'm wondering what they are referring to here - what portable systems have 10Gbps ports?

On a related note, several top tier network engineers I work with who do a lot of work at Layer3 (Writing Mesh routing protocols) - say if they could have one, and only one book for the rest of their life, it would be Radia Perlman's "Interconnections, 2nd Edition." I'll attest that some chapters have captured my attention for months. :-)

> Fast Ethernet uses the same CDMA/CD as Ethernet

> That should be CSMA/CD. They had it right earlier and later on, not sure how this reference got by them

It's a one letter typo. The keys are adjacent on the keyboard. Enhance your calm.

My angst with the typo, is that I work on CDMA and GSM networks all day. Now, I know that it really should mean CSMA, but can you imagine someone from a different context reading this and wondering how CDMA and Ethernet tie together?

Regardless - less of an issue on a blog post, but Ars normally sets a really high bar for copy editing - odd they let that one get by them.

Could also be a moment of lysdexia. (In acronym, no spell checker hears you scream.)

Not necessarily networking related, but it is a 10Gbps port: http://www.intel.com/technology/io/thunderbolt/index.htm

"Think about it: is there any other 30-year-old technology still present in current computers? 300 baud modems? 500 ns memory? Daisy wheel printers?"

IBM Model M keyboards.

Not to mention tons of ubiquitous algorithms.

Not just the technology's design, but the actual devices were so well made, their function is often still just fine after 30 years, and they're still plugged into recent computers.

my favorite part of the article:

Algorhyme

        I think that I shall never see
        a graph more lovely than a tree.
        A tree whose crucial property
        is loop-free connectivity.
        A tree that must be sure to span
        so packet can reach every LAN.
        First, the root must be selected.
        By ID, it is elected.
        Least-cost paths from root are traced.
        In the tree, these paths are placed.
        A mesh is made by folks like me,
        then bridges find a spanning tree.

                         Radia Perlman

I was liking it too, until the abrupt meter change in line 7. People almost invariably mess up the meter when writing this kind of verse. It's a hallmark of tone-deafness, or whatever the equivalent is for rhythym. For anyone who is paying attention, it spoils the effect.

But then I noticed something surprising. Yes, the meter changes abruptly from iambic to trochaic at line 7, i.e. the stress goes from even syllables ("I think that I shall never see") to odd ones ("First the root must be selected.") However, the shift is executed perfectly: the meter is a steadfast trochee for four lines and then reverts to iamb for the final couplet.

Accident? I doubt it. Each of the four trochaic lines is a complete sentence. The iambic lines all have exactly one sentence per two lines. No, this is the rare case where the writer knows exactly what she is doing. Bravo Radia Perlman!

Well spotted. Also, the four trochaic lines list the steps of the tree-building process, and the iambic couplets explain the purpose of it all. She's using meter to reflect a semantic division.

You may also like this then:

http://web.mit.edu/adorai/www/seuss-technical-writing.html

While the article says that deciding if a crossover cable is needed is still a problem today, in my experience devices figure out what cable they have attached and just work.

When will we see consumer devices rated at above 1gigabit?

It comes down to there being logical use case scenarios for it.

Most computers can't write to disk at 1Gbit/s, and the power and related infrastructure (multiple PCIe lines) required for a 10Gbit Ethernet transceiver is higher than for a gigabit one.

When some application comes along that is greatly improved by 10 gigabit networking (rather than point to point like Thunderbolt, USB3, or SATA), then we'll see it happen.

It's hard to predict; 10G prices have plateaued or increased in recent years, so maybe the answer is never or maybe there will be a dramatic price drop some day.

This Summer: http://www.lacie.com/us/products/product.htm?id=10549

That is not an ethernet device. Consumer devices have had non-ethernet interconnects faster than 1 Gbps for a while now. (eSATA, USB 3.0, lots of internal interconnects such as PCIe.)

Why do consumers need such a thing?

There are a few places you can get gigabit fiber, even excluding Google's offering.

How does this explain the need for greater than gigabit networking?

Netflix