The time lady!
At the tone the time will be One Fifty Nine and Forty Seconds. *BEEP* At the tone the time will be One Fifty Nine and Fifty Seconds. *BEEP* At the tone the time will be Two O'clock Exactly. *BEEP*
150MB? That's spacious. I have a 10MB MFM drive in the garage with a big ol' stepper motor on it to move the heads. 80ms seek time, baby! Zzzzt...zz..zzzt...zzzt...ztztztztzt....
Hearing the opening music over the tape loading sounds reminded me of this track, involving Pirate Adventure for the TI-99/4A.
I never experienced a mimeograph, but I do remember the sound (and smell!) of the Ditto machines (spirit duplicators).
As for a different sound that's largely died out: Fast forwarding / rewinding / ejecting a cassette tape.
Well, in this case the road was "clear" as the snow did melt immediately (road was till to warm) nevertheless the risk of "ice" here and there was given. Imagine snow melting and freezing a few yards further again.
I'm all too familiar with that, having grown up in snow. We called that slick ice that results from melting and refreezing "black ice." It was especially bad in areas with wind blowing snow over the road, as that snow would obscure the ice.
Lane detection on a highway is easy as you have a barrier on one side and poles on the other, actually you have poles on both sides.
Ah. Many of our highways here lack barriers and poles, instead just using large grassy areas to divide the highway from the surrounding environment, such as this. Those same highways can be up to 3 lanes wide. I've driven on I-94 (recently the site of a 90-car pileup) in conditions where drivers created 2 lanes out of the 3 that exist, because there wasn't any real reference to go by to find the lanes.
Self driving cars rely on a sensor array (short-range RADAR, long-range RADAR, LIDAR, SONAR, stereo and infrared cameras, temperature and accelerometer feedback) to build their world model. I admit I'm not a sensor expert myself, so I'm not really sure how degraded those sensor inputs would be in fog and snow. Obviously the thermometer and accelerometer would be fine, but each of the others would experience degraded range.
The self-driving car may end up going a bit slower than 80km/h to avoid out-driving its ability to look ahead. Depth computed from stereo vision I imagine would be heavily impacted, as fog isn't completely uniform. I'm not sure how LIDAR would perform. RADAR may be OK, as we're able to track planes through clouds just fine.
I also wonder how lane detection even has a chance in snow, unless the roads have been cleared, or there are clear lane ruts from previous cars. Having grown up in snowy Michigan, I know human drivers often get the lane boundaries wrong after a heavy snow.
Ah, that makes sense. Inclement weather is 2x - 3x as stressful to drive through, and certainly much more dangerous. I doubt an autonomous car would have been great in those conditions either, though.
I guess I was being a bit snarky earlier.
For some things, I can see waiting to package it up at the post office. Priority / Express Mail envelopes are a big one. Since I do a fair bit of shipping via USPS, I actually go down to the PO ahead of time and pick up boxes for the things I'm going to ship, and do all the packaging at home. That's how I'm acutely aware that they often don't have what I need at a particular location, which further reinforces my desire to pack things up at home.
I also don't trust the tape-strips that are pre-applied to the packaging. I've had some that stuck like glue, and some that started peeling open a minute after packing up my boxes. So, I always end up reinforcing with some shipping tape.
I grant you, I'm probably not the typical postal customer, though. The more casual user of postal services will probably scratch their heads over many of the things you've pointed out on the USPS site. I've been using it for so long, it all feels pretty natural to me.
I imagine the "no first class stamps at USPS.com" thingy is a concession to Pitney Bowes, Stamps.Com and other vendors they've made deals with for various first class postage services. Otherwise, there'd be little reason for those other services to exist.
You could argue that it's the user's responsibility to make sure their package fits into the box they select, but a user could reasonably assume that the whole point of entering the length, width and height is so that the USPS can recommend only those boxes that will hold the item. Remember, the user usually doesn't have these boxes in front of them at the time they're printing the label. They could end up selecting a box option, printing the label, taking it all the way to the post office along with their package, only to find out that the package doesn't fit into the box that they printed the label for, and that they have to wait in line anwyay to pay for an alternate method.
Ah, you're one of those people who clog up the lobby boxing your stuff up at the post office, using the wrong tape (such as the tape meat to mark an Express package on something you're shipping Priority or First Class) and breaking in line to ask someone behind the desk for scissors.
You realize that the post office isn't a full service pack and ship place, right? At least none of the ones I've been to around here are. You're supposed to have everything packed up and ready to go before you walk in the door. You also realize that your local PO probably doesn't stock all the sizes and shapes of shipping box the website describes, and that package weight is supposed to include the box, right?.
That is, you're supposed to have boxed up your parcel by the time you got to this part of the form. The only thing missing should be the label.
Could be worse. You could be like the person I saw who tried to send a package wrapped in normal Christmas wrapping paper.... That was going to be a shredded nightmare on the other side.
On some Linux distros,
The only other case I can think of where a RAM drive might make sense is if you have a set of files you need access to with tight deadlines, and the total corpus fits in RAM. Of course, you could also mmap and mlock those files to hold those files in RAM, if you have control over the application's implementation. For example, in the bad old days of 4x CD burners with almost no buffer, loading the ISO into a RAM disk could help weak burning software keep up with its realtime deadlines. That is, if you had enough RAM to hold the ISO (it needed to be a smallish ISO, not a full 650MB).
Otherwise, RAM disks are usually a bad idea these days.
You needed to stay overnight somewhere for a mere 6hr drive? I usually reserve that for drives over twice that long.
LIDAR stands for Laser Infrared Detection and Ranging. Why does the summary say "(light, radar)" after LIDAR? RADAR uses radio waves, not infrared laser.
(And yes, Mr. Pedantic, I realize radio waves and infrared light waves are both electromagnetic waves. But, our mechanisms for detecting things in the radar band vs. the infrared light band are quite different, so the distinction is meaningful.)
Eventual consistency means that the computer eventually computes the right answer if its quiescent long enough. Intermediate values, though, are an approximation, which is often enough.
One example that Paul McKinney gives is of a distributed counters built out of per-CPU counters, and CPU-to-CPU events saying how much to update the total by. (Let's assume positive counts only.)
Each CPU will see update events from other CPUs in different orders, each saying how much to update the count by. All CPUs will eventually see all updates. So, the total seen by any given CPU might differ from the true total in the short run (and may not even be a technically valid total given the original source of events, since events get reordered), but eventually all of the counters will converge on the same total if updates stop pouring in. Also, the totals are still locally monotonic.
If you required all CPUs to see the same sequence of updates to the count, then you have to take locks and serialize memory accesses, which on a manycore system is an expensive operation that simply doesn't scale well. But, if you relax the constraint to "eventual consistency" and "monotonic updates", then each core can have its local approximation that isn't too far from the real value, knowing that each core is no further from the true value than the backlog of events yet to arrive.
That's an extremely reasonable model for many types of data.
Go buy some cheap hosting and set up a WordPress blog then. You can have that nice, shiny island on the Internet all to yourself just like you like it. Tell all your friends to subscribe to its RSS feed.
Neutrinos have bad breadth.
