Considering the.ca system is regularly overloaded -- to the point where OpenSRS implemented a special 'delayed transaction' system for it -- it's not that good.
The Affilias.org (and.info) systems get the domain up and running in under 2 minutes -- which is really good considering a good chunk of that time is DNS propogation related (between masters for the domain).
Thats just about guarenteed to happen. There is an XML or XML related standard change about once a month -- and that doesn't include the specific protocol implementations using XML.
I'm afraid that there hasn't been a big effort to make replication user friendly -- whats there works very nicely as asynchronous master -> multiple slaves.
There are a few things on the developers minds prior to replication. 2 Phase commits, improved inter-database connections (see dblink), Point in Time Recovery (via WAL logs).
Once any 2 of the above three are completed, then replication should be a piece of cake:)
That said, I'm holding out on asynchronous multi-master replication.
Well... thats half true. Certainly more bandwidth, but not normally lower latency.
Light through vacuum is quick, light through glass isn't as quick. Couple this with the inability for the light to travel in a straight line through the fibre (it bounces around off the sides -- more or less). Electrical signals through copper don't experience these affects as much.
Lastly, a rather complex and heavily delayed circuit has to convert the electrical signals to light, and back again. This takes time -- but the percentage of time taken is small in comparison to normal travelling distances but don't expect them to make a slowed down PCI bus using fibre any time soon.
The only way to speed up harddrives is to pass more bits infront of the drive head in a set amount of time.
Add more platters and/or readheads, spin them faster, or compress the bits so that more pass per revolution as more fit in the same space.
Since anything faster than 10k seems to heat up in a hurry you won't find them in a home system soon. Nor will you find 'large sized' drives soon. Good chance platters could become thinner, and put more into the housing but thats an expensive proposition. Data compression (physical, not mathematical) like IBM is doing is a very effective method of complying with your request.
I realize the article says the universal charger is under development. The big problem is failure. When it mis-communicates, you don't drop a packet and retry. Odds are you're device will fry as the power supply may set itself to too much or too little a voltage level causing circuits to fail.
Remember back to the 80s and how clunky things seemed to be? Most of that was due to the battery. Everyone used the same batteries to help bring down costs, but then the final design was around the battery.
Since formed batteries have dropped in price significantly, they can make nice sleek designs and work the battery in rather than the other way around.
Whats this have to do with your question? If everyone could use the same power source, it would mean all devices would be required to support N Volts at N Amps (or set internal resistance accordingly). This means the addition of a step up or down DC-DC transformer to all devices, and taking a step backwards on battery design as you must accommodate the lowest common demoninator again.
However, not all is lost. There is a slim possibility you could add a layer of communications with the block and request a specific voltage from it, thus placing the logistics into the power block.
Figure out how to do that in a small package, and it may simply become the AC outlet. Quite clunky to accomodate at the moment, so specific individual bricks rather than an über brick is far cheaper, smaller, and much less prone to failure.
For now, just be glad the brick isn't an integrated part of the device.
Yes, sorry, broadcasting is an entirely different matter, as you normally cannot broadcast anything without the persons knowledge and consent. There are special provisions for news worthy events but I've never really looked into it.
Unsure where you are, but if you leave your blinds open you can expect to have zero privacy. Neighbours and others are well within their right to watch and record anything you do within your home.
If you close your blinds, you can expect privacy. This is law enforcement authorities require a warrant for microwave or infrared monitoring, but standard video cameras don't. The general public can't see what you would be doing, so under general circumstances neither than the authorities.
Now.. If you leave your computer open with full read access, I'd say you forgot to shut the blinds and can expect the privacy that goes along with it. Reading your email at a public terminal certainly doesn't grant you rights to privacy. You've used absolutly no precautionary rights.
Bandwidth and sales (or general broadcasting) of such material may have cases, but the fact you were allowed to read the data means you're allowed to read the data.
In summary, don't read those all important and secret corporate financial reports on a crowded subway. Those are you have the right to read them as well, regardless of whether you consider it rude to read over anothers shoulder.
Well.. Some databases don't lock rows or tables (unless necessary). MVCC is a superior option (Oracle / Postgresql for implementation references).
Anyway, there are a number of things that subselects can accomplish that joins cannot do as easily. Not the best example below, as it could be done in other ways and the formatting isn't so great, but anyway:
SELECT col1, max, othercol FROM table JOIN (SELECT max(col3) as max
, col1
FROM table3
GROUP BY col1) as ttab
USING (col1) WHERE 2 = (select count(*)
from table2
WHERE table.col1 = table2.col1);
Slashdot Mirror
Considering the .ca system is regularly overloaded -- to the point where OpenSRS implemented a special 'delayed transaction' system for it -- it's not that good.
.org (and .info) systems get the domain up and running in under 2 minutes -- which is really good considering a good chunk of that time is DNS propogation related (between masters for the domain).
The Affilias
Thats just about guarenteed to happen. There is an XML or XML related standard change about once a month -- and that doesn't include the specific protocol implementations using XML.
2) And, if you add the SAME number to the beginning of everything, that gives you nothing. Why would they do that?
Actually it does give you something. It will allow you to use 0 or 1 as the second digit -- thus *buying* 2 billion more phone numbers.
Of course, globally routing numbers (drop the concept of 'area code', and just make it 3 arbitrary numbers) would do more for the system.
We won't know what we will learn until we get there -- much as we didn't know what we'd learn on the moon until we got there.
Yes, we did learn *a whole bunch* by going to the moon, even if most of it wasn't evident until recently (technological gains).
By going to Mars, I'll be looking a few decades later for another kevlar, microchip, or similar coming out of it.
Really, what we learn from mars won't be so big. What we learn from the trip itself could be huge.
Actually, the trick is to do it without comitting cannibalism.
See the last paragraph where it's briefly discussed that one may possibly be able to grow a steak out of a self biopsy.
Still grossed out? I'm not.
Well, here I am.
Sounds awfully Boonian to me.
Thats exactly what it means..
SELECT * FROM function()...
Not to mention in PostgreSQL almost all DDL statements (like Drop Column) are transactional -- which adds a touch of added complexity.
I'm afraid that there hasn't been a big effort to make replication user friendly -- whats there works very nicely as asynchronous master -> multiple slaves.
:)
There are a few things on the developers minds prior to replication. 2 Phase commits, improved inter-database connections (see dblink), Point in Time Recovery (via WAL logs).
Once any 2 of the above three are completed, then replication should be a piece of cake
That said, I'm holding out on asynchronous multi-master replication.
Well... thats half true. Certainly more bandwidth, but not normally lower latency.
Light through vacuum is quick, light through glass isn't as quick. Couple this with the inability for the light to travel in a straight line through the fibre (it bounces around off the sides -- more or less). Electrical signals through copper don't experience these affects as much.
Lastly, a rather complex and heavily delayed circuit has to convert the electrical signals to light, and back again. This takes time -- but the percentage of time taken is small in comparison to normal travelling distances but don't expect them to make a slowed down PCI bus using fibre any time soon.
The only way to speed up harddrives is to pass more bits infront of the drive head in a set amount of time.
Add more platters and/or readheads, spin them faster, or compress the bits so that more pass per revolution as more fit in the same space.
Since anything faster than 10k seems to heat up in a hurry you won't find them in a home system soon. Nor will you find 'large sized' drives soon. Good chance platters could become thinner, and put more into the housing but thats an expensive proposition. Data compression (physical, not mathematical) like IBM is doing is a very effective method of complying with your request.
Yup.. That was my first week with unix. vi is tough to quit, but I never once anticipated :help being the way to help.
:)
Thanks to ee I seem to have made it
Vim truely is an improvement on vi. Tells you how to exit the first time you accidentally load it.
I realize the article says the universal charger is under development. The big problem is failure. When it mis-communicates, you don't drop a packet and retry. Odds are you're device will fry as the power supply may set itself to too much or too little a voltage level causing circuits to fail.
But, I'd like them to proove me wrong.
Remember back to the 80s and how clunky things seemed to be? Most of that was due to the battery. Everyone used the same batteries to help bring down costs, but then the final design was around the battery.
Since formed batteries have dropped in price significantly, they can make nice sleek designs and work the battery in rather than the other way around.
Whats this have to do with your question? If everyone could use the same power source, it would mean all devices would be required to support N Volts at N Amps (or set internal resistance accordingly). This means the addition of a step up or down DC-DC transformer to all devices, and taking a step backwards on battery design as you must accommodate the lowest common demoninator again.
However, not all is lost. There is a slim possibility you could add a layer of communications with the block and request a specific voltage from it, thus placing the logistics into the power block.
Figure out how to do that in a small package, and it may simply become the AC outlet. Quite clunky to accomodate at the moment, so specific individual bricks rather than an über brick is far cheaper, smaller, and much less prone to failure.
For now, just be glad the brick isn't an integrated part of the device.
Yes, sorry, broadcasting is an entirely different matter, as you normally cannot broadcast anything without the persons knowledge and consent. There are special provisions for news worthy events but I've never really looked into it.
Ok.. Checked, FreeBSD 4.7 required reboot anyway.
Bios version 2.21 with a date of 2001-10-19
Unsure where you are, but if you leave your blinds open you can expect to have zero privacy. Neighbours and others are well within their right to watch and record anything you do within your home.
If you close your blinds, you can expect privacy. This is law enforcement authorities require a warrant for microwave or infrared monitoring, but standard video cameras don't. The general public can't see what you would be doing, so under general circumstances neither than the authorities.
Now.. If you leave your computer open with full read access, I'd say you forgot to shut the blinds and can expect the privacy that goes along with it. Reading your email at a public terminal certainly doesn't grant you rights to privacy. You've used absolutly no precautionary rights.
Bandwidth and sales (or general broadcasting) of such material may have cases, but the fact you were allowed to read the data means you're allowed to read the data.
In summary, don't read those all important and secret corporate financial reports on a crowded subway. Those are you have the right to read them as well, regardless of whether you consider it rude to read over anothers shoulder.
I've been running FreeBSD on my x20 for a little over 5 months now -- Type 2662.
I made a converter between laptop and standard desktop IDE cables, so that I could format the drive if it failed to boot properly -- but it did.
Install the most recent BIOS & controller upgrades and go to it.
Still doesn't help much when the person you're talking to is the informant ;)
There are 10 kinds of people in the world. Those who understand binary, and those who dont.
Well.. Some databases don't lock rows or tables (unless necessary). MVCC is a superior option (Oracle / Postgresql for implementation references).
Anyway, there are a number of things that subselects can accomplish that joins cannot do as easily. Not the best example below, as it could be done in other ways and the formatting isn't so great, but anyway:
SELECT col1, max, othercol
FROM table
JOIN (SELECT max(col3) as max
, col1
FROM table3
GROUP BY col1) as ttab
USING (col1)
WHERE 2 = (select count(*)
from table2
WHERE table.col1 = table2.col1);
How exactly does a lock help rollback the data change when the database crashes -- an running by naked trips on plug -- it's happened...
Not only that, but they keep getting investments because of their potential.
Of course, any decent regression test suite ensures the negative cases fail as often as the positive cases pass.
You can't test simply one subset of the API.
Lol.. I just realized there is another Starbucks and Tim Hortons underground, one story below street level.
So that makes 5 shops within about 150 paces.