Forgot your password?
typodupeerror

Comment: Won't Last Long (Score 1) 124

by herbierobinson (#48177565) Attached to: Federal Government Removes 7 Americans From No-Fly List

The actual announcement has this phrase in it: "... the following individuals are not currently on the No Fly List as of the date of this letter". They explicitly go on to state that "we make no other representations with respect to past or future travel". In other words, as soon as the court case is over, they go right back on the list...

Comment: Re:Waaah. (Score 1) 338

by herbierobinson (#47743157) Attached to: New EU Rules Will Limit Vacuum Cleaners To 1600W

Devices designed for standard outlets in the US aren't allowed to draw more than 12 amps continuous. At 120V (the standard voltage), that works out to 1440 Watts. One will find that many things are rated at that wattage, because one has to use a different plug to go beyond that. The outlet circuits are generally fused at 20 amps to allow more than one device to be plugged in (and there can be more than one outlet on the 20A circuit).

Comment: Re:Thanks (Score 1) 398

Large window sizes only help when there is absolutely no packet loss (which is not the Internet)
and even then it takes a long time to scale up.

When packet loss is significant, it will indicate congestion or a network issue and be a more important factor than latency.

Packet loss and latency are both significant. The recovery time from a single packet loss is proportional to the latency.

Large windows still help. Packet losses are usually less than 0.01%. Anything above approximately .1% is a strong indication of a possible issue.

Large windows should still help; although, there might be some rare situations where they make performance worse (if the network they are on is saturated and the extra traffic pushes it past a tipping point so packet loss increases).

When one is dealing with large windows, even 0.01% packet loss hurts. This is because it takes longer to get back up to full speed after a packet loss. It will still be faster than a smaller window size, but the packet loss will eat away more overall throughput than it would with a smaller window size.

Often websites are bypassing slow start by starting with an expanded initial congestion window.

Yes, if they are following the RFCs, they can use up to 4 packets (assuming the MTU is small) for the initial congestion window. That cuts two RTTs off the ramp up time (very roughly 10%). The last time I checked, web sites were usually transferring less than 100K bytes -- The use of 4 packets for the initial congestion window should actually help them a lot more than having a large window size (that they will never actually ramp up to). Where the large window sizes really help are large uni-directional transfers (ftp, sftp, etc).

Comment: Re:Thanks (Score 1) 398

Large window sizes only help when there is absolutely no packet loss (which is not the Internet) and even then it takes a long time to scale up. Even when there is a large window size, the slow start mechanism is really really slow when the RTT is in the 100 msec range (the Internet). If you have regular low level packet loss, the actual throughput plot looks like a sawtooth wave and the upward ramp is very slow. The congestion window rarely even gets close the max allowed by the window size in the TCP header. It can take several minutes for the congestion window to ramp up with 100msec latency.

SACK might help a little bit if implemented properly, but even SACK is still required to do the slow start congestion avoidance.

There are RFCs (I haven't had time to read, yet) where routers get involved with congestion control. But the RFCs haven't become standard, yet, and the chances that every router in an Internet path supports them is probably negligible.

Comment: Too Epensive (Score 1) 190

The kind of cryptography you would need to make this safe (as posted earlier) is Public Key. The big problem with that is teaching ALL the voters how to use it. And we aren't even talking about distributing the software or handing out keys: That's easy compared to the training...

Comment: Re:What of the downstream bandwidth usage? (Score 1) 390

by herbierobinson (#47495287) Attached to: Verizon's Accidental Mea Culpa

Verizon's analysis would indicate that's not the case, unless the Netflix traffic is multicast, but the traffic is on demand; so, I kind of doubt much of it is multicast.

Even if it is, Verizon's customers are paying for for bandwidth at the edge of the network; so, they should be getting internal network support for that.

Comment: Not So Obvious (Score 1) 390

by herbierobinson (#47495205) Attached to: Verizon's Accidental Mea Culpa

We don't have all the information here. The Level 3 post only talks about how much the hardware costs.

Level 3 very explicitly avoided mentioning services charges which Level 3 may (or may not) have been trying to get from Verizon. Note that Level 3 didn't explicitly say they would provide the extra connection bandwidth free of charge, only the hardware. I think there is more going on here than we know...

Of course, it could also be that Verizon is trying to get money from Level 3 because Level 3 was sending so much more traffic than Verizon. I am a very experienced network software designer and it is more expensive to receive IP traffic than to transmit it -- especially non-TCP traffic, because there is no flow control -- you need a lot of extra hardware capacity to handle bursts (buffering and CPU). From a technical standpoint, it makes a lot more sense for Level 3 to pay Verizon to handle the extra traffic.

BUT, compared to the cost of the entire network, the real cost of the peering is probably pretty insignificant for both parties. The only conclusion here is that they are probably both spinning the message by leaving out the unflattering information.

Comment: What Operating System? (Score 1) 383

by herbierobinson (#45851167) Attached to: Ask Slashdot: Command Line Interfaces -- What Is Out There?

Every operating system I know of (except maybe BeOS) has at least one command line interface. Many operating systems have more than one.

For example, the Stratus VOS operating system supports this command line interface:

http://stratadoc.stratus.com/vos/17.2.0/r089-06/wwhelp/wwhimpl/js/html/wwhelp.htm?context=r089-06&file=ch1r089-06.html

This command line interface was designed around 1980 and attempts to be more user friendly by using recognizable command names and has both lineal and form oriented methods for specifying command arguments.

The VOS operating system also supports bash...

Comment: Re:Missing the point (Score 1) 668

by herbierobinson (#45054915) Attached to: Are Shuttered Gov't Sites Actually Saving Money?

Technically, it's not unconstitutional. It's been thought of before: Many years ago, I asked my Congressman why the House didn't block funding the 2nd Iraq war (the Dems had a majority at the time) and the response could be paraphrased that while it was possible, it simply wasn't done, because in the long run we need to put the country's interest in front or our own. Other than that, you have it spot on.

What's happening now is the rich .1%ers who actually pay for the Tea party don't want to pay their fair share of taxes (which eventually has to happen to pay for health care -- irrespective of whether you are talking about Medicare, Medicaid or Obamacare) and they simply don't care what happens to the rest of us.

Comment: Re:Slippery slope. (Score 1) 604

by herbierobinson (#43512607) Attached to: Bruce Schneier On the Marathon Bomber Manhunt

Exactly.

Dont forget the lockdown was voluntary. One would have to have been pretty nuts to open a business in that scenario, but nobody was actually forced to stay home. I live in the locked down area and I was half way to work before I even heard about it. The only effort they actually made to contact ordinary citizens was via press releases and an e-mail (which in my case ended up in the junk mail account -- didn't find it until I got home that night and looked for it). Of course, I was a good 5 miles from where anything was actually happening.

They did announce the end of the lockdown with two robocalls. I suspect they did get a lot of complaints.

Comment: Re:A good reason to go independent (Score 1) 550

by herbierobinson (#40967521) Attached to: Is Your Neighbor a Democrat? There's an App For That

But party membership isn't private (at least in Massachusetts). Anyone can join a party by going to the town/city clerk and registering as a member of the party (as simple as checking a checkbox when you register to vote). All of the meetings are open to the public. Membership in the Ward/Town/City/State committees is voted on in the primaries (although, it comes with a time commitment; so, it's often not contested). It's not some secret club that nobody can belong to. The reality is that participation is low because most people are too lazy to get involved.

What's not open is how candidates get money (for the most part, that's not from the party).

Blessed be those who initiate lively discussions with the hopelessly mute, for they shall be known as Dentists.

Working...