1) Install Wikipedia software locally and use this for any locally created articles
2) The web server running this simply proxies out to en.wikipedia.org for that request if not available in the local version. The easiest way to do this is with Apache + rewrite rules
This means that users can get to articles locally and on wikipedia from the same command
You then need to consider the following
1) The search request needs to go to the local version of wikipedia then the external one and concatinate the results together - a small proxy script should be more than capable of doing this
2) You may want to create a reference table which maps external wiki articles to related internal ones. Again a small script could insert these into the external wikipedia articles during rendering
If you are running a bog standard desktop or laptop you probably have some basic requirement * when I turn up at my computer after having a cup of coffee and hit slashdot.org into firefox I expect it to launch straight away * when I shutdown my computer I expect it to happen instantly * when I start up my computer I expect it to happen instantly * when I save my music/words/picture I expect it to be reliable and nothing my computer will do after I save my file will affect it * under no circumstance will my mouse, keyboard or icons on desktop not perform to my actions - I want no hang.
This is not the fault of disk swapping it is just an example about how 5 utterly straightforward requirements have not been addressed in over 20 years of desktop development.
If your servers are operating at more than 25C on a regular basis then chances are you are seeing high rates of disk drive failure and other component based failures. I can't imagine the downtime, cost of replacement parts and effort to fix would be cheaper than spending the $600.
We had an AC outage for just 1/2 an hour on about 12 racks. Over the nexct 3 months lots of random disk failures cropped up in that area.
Isn't it about time that ISP's were upfront and simply charged users for what they use? This would encourage ISP's to grow bandwidth to meet demands ( as it adds revenue ) and for users to decide how much content they wish to pay for
10,000,000 * 7MB file over 24 hours = 8gbit/sec sustained or 16gbit/sec peak.
Small fry for a CDN but pretty hefty for anything centrally hosted. Although in this case it looks like they a DNS mapping to mirror servers around the world.
First off India and Turkey add up to a huge number of Muslims able to participate in democracy. Second Islam is not a religion that encourages superstition. Third look around you at any scientific institute in America, UK and beyond and see the vast contribution Muslims make to modern day research.
I aint Muslim but am getting a bit fed up of the continous bashing they get while working alongside them.
Last time I checked you create Index's on columns which are regularly accessed in a serial manner for lookups
And you make a balance between the number of index's you need versus the type of queries to be done ( more index's = faster lookup but slower row by row insertion )
Does sound like Column Store Database is a subset of the functionality of a half decent Relational Database
Thankfully in linux there is one person who can make a clean decision between 2 competing but adequate parts of kernel. The idea that a core part of Linux could be switched between would mean that any product worth its salt would need a test regime that would double in complexity to deal with systems that chose to 'switch' from one to the other.
Expecting any guide to provide the definitive perfect global view of topics like 'Jesus', 'Islam', 'terrorism' is asking a bit much. In fact we might find the endless edits and discussion on these key topics provide invaluable data to future anthropologists for the Zeitgeist around a topic at a particular time.
Any book or guide purporting to be the definitive guide would be the anti-thesis of many social science aims - we should encourage skepticism of sources. Like any secondary or tertiary source - as long as you use with open eyes then no problem.
While the idea needs some more work if it saves this volume of energy it is worth serious investigation. I am afraid the only 'phenomenally stupid idea' is having 100 million appliances which need to be working at full pelt for no other reasons than the way the software on them is designed.
Imagine the laughs if a new car was brought out which required the engine to be on all the time - because if you turned it off you cannot unlock the doors.
Little more complex than ensuring network hardware is up to it. There is a lot of applications which think an ip address is a group of 4 numbers from 0-255 separated by dots.
Bit like year 2000 - going to be very hard for organisations to check all apps can cope.
Might be good check on the dependancy on Unix Epoch time while doing scanning applications
Upgrades are always a nightmare - some thoughts
on
Paypal Grinds To A Halt
·
· Score: 2, Insightful
1) You should always have a detailed rollback plan. This should include what you monitor during the change, what triggers the rollback and who is responsible for making that decision
2) You should have a rollback plan for any change to a live site which deals with transactions no matter how 'small' the site is
The problem these days is the complex interaction between different systems and user traffic means there is no guarantee of a flawless rollout even with a full staging system to test on. I have seen failures occuring due to a unique combination of user requests creating a thread contention which results in an instability causing failure up to 1/2 an hour later!
There is a gap for a system which can capture the entire network traffic being applied to a system and replay it against a staging environment to catch just this sort of problem.
The other problem is a 'minor' problem during rollout such as intermittant failure of some parts. This is let go for a while and people start throwing hacks at the problem. Before you know it enought new transactions have been applied to the system to make rollback unviable.
In my opinion with complex web based systems testing and rollout is becoming one of the most interesting areas in IT at the moment
Just had a quick look at the tarball. It is a 22 MB substitution phrase based DB read by a reasonably simple PHP script.
It seems to have basis ability to correctly position Proper-Nouns using wild card characters within phrases.
No clever grammar rules etc which is probably a good thing. Stick on a 'did this translate properly' button and let users add to the vocabulary is probably a better approach long term approach with enough users that a clever grammatical algorythm.
Paxman has a huge knowledge of modern politics and politicians which he has used to devastating effect on numerous British politicians in the past
I felt he wasn't as confident or as dangerous with Bill Gates through a lack of technical knowledge.
It is a crying shame than interviewers of the calibre of Paxman rarely have the technical background to deal confidently with these goliaths of IT as they have so succesfully done with politicians.
Slashdot Mirror
I had 3 backups of home data of about 300gbytes each.
Each one was almost but not quite the same due to some rather poor backup policies on mypart.
I was able to dedup per backup to get them small enough to combine and dedup the combo.
Left with one pure 150gbytes combo. Rsync is amazing
Only problem is getting your changes into the trunk.
For that I am glad they are very picky.
Ensuring everyone in project is using same one is more important than which one is chosen.
If you can use the same one for a whole company even better
1) Install Wikipedia software locally and use this for any locally created articles
2) The web server running this simply proxies out to en.wikipedia.org for that request if not available in the local version. The easiest way to do this is with Apache + rewrite rules
This means that users can get to articles locally and on wikipedia from the same command
You then need to consider the following
1) The search request needs to go to the local version of wikipedia then the external one and concatinate the results together - a small proxy script should be more than capable of doing this
2) You may want to create a reference table which maps external wiki articles to related internal ones. Again a small script could insert these into the external wikipedia articles during rendering
If you are running a bog standard desktop or laptop you probably have some basic requirement
* when I turn up at my computer after having a cup of coffee and hit slashdot.org into firefox I expect it to launch straight away
* when I shutdown my computer I expect it to happen instantly
* when I start up my computer I expect it to happen instantly
* when I save my music/words/picture I expect it to be reliable and nothing my computer will do after I save my file will affect it
* under no circumstance will my mouse, keyboard or icons on desktop not perform to my actions - I want no hang.
This is not the fault of disk swapping it is just an example about how 5 utterly straightforward requirements have not been addressed in over 20 years of desktop development.
First OS which has this gets my vote
If your servers are operating at more than 25C on a regular basis then chances are you are seeing high rates of disk drive failure and other component based failures. I can't imagine the downtime, cost of replacement parts and effort to fix would be cheaper than spending the $600.
We had an AC outage for just 1/2 an hour on about 12 racks. Over the nexct 3 months lots of random disk failures cropped up in that area.
Isn't it about time that ISP's were upfront and simply charged users for what they use? This would encourage ISP's to grow bandwidth to meet demands ( as it adds revenue ) and for users to decide how much content they wish to pay for
10,000,000 * 7MB file over 24 hours = 8gbit/sec sustained or 16gbit/sec peak.
Small fry for a CDN but pretty hefty for anything centrally hosted. Although in this case it looks like they a DNS mapping to mirror servers around the world.
Available today for under $500
Runs Linux so no need for chunky processor - doddle to use and no moving bits like Harddrive or DVD drive to break.
Is an option at a couple of local independant PC store near me
From Vista of course. Costs £50 a pop and they guarentee not to wipe any of your data - have your computer 'fixed' in under 24 hours.
First off India and Turkey add up to a huge number of Muslims able to participate in democracy. Second Islam is not a religion that encourages superstition. Third look around you at any scientific institute in America, UK and beyond and see the vast contribution Muslims make to modern day research.
I aint Muslim but am getting a bit fed up of the continous bashing they get while working alongside them.
Last time I checked you create Index's on columns which are regularly accessed in a serial manner for lookups
And you make a balance between the number of index's you need versus the type of queries to be done ( more index's = faster lookup but slower row by row insertion )
Does sound like Column Store Database is a subset of the functionality of a half decent Relational Database
Thankfully in linux there is one person who can make a clean decision between 2 competing but adequate parts of kernel. The idea that a core part of Linux could be switched between would mean that any product worth its salt would need a test regime that would double in complexity to deal with systems that chose to 'switch' from one to the other.
Of course controversial subjects are fought over.
Expecting any guide to provide the definitive perfect global view of topics like 'Jesus', 'Islam', 'terrorism' is asking a bit much. In fact we might find the endless edits and discussion on these key topics provide invaluable data to future anthropologists for the Zeitgeist around a topic at a particular time.
Any book or guide purporting to be the definitive guide would be the anti-thesis of many social science aims - we should encourage skepticism of sources. Like any secondary or tertiary source - as long as you use with open eyes then no problem.
While the idea needs some more work if it saves this volume of energy it is worth serious investigation. I am afraid the only 'phenomenally stupid idea' is having 100 million appliances which need to be working at full pelt for no other reasons than the way the software on them is designed.
Imagine the laughs if a new car was brought out which required the engine to be on all the time - because if you turned it off you cannot unlock the doors.
Not the case bbc.co.uk has been showing certain Comedy Series 7 days before they are broadcast. They include 'Mighty Boosh' last autumn
The 'Empty Child' in particular. I remember hiding behind the Sofa 20 years ago during the earlier series as this one induces that reaction now.
Little more complex than ensuring network hardware is up to it. There is a lot of applications which think an ip address is a group of 4 numbers from 0-255 separated by dots.
Bit like year 2000 - going to be very hard for organisations to check all apps can cope.
Might be good check on the dependancy on Unix Epoch time while doing scanning applications
I hate that expression.
In fact if you compare
& q= kate+winslet&rys=0
T F- 8&fr=sfp&p=kate+winslet
http://www.alltheweb.com/search?cat=vid&cs=utf8
http://video.search.yahoo.com/search/video?ei=U
You will see remarkable similiarities
1) You should always have a detailed rollback plan. This should include what you monitor during the change, what triggers the rollback and who is responsible for making that decision
2) You should have a rollback plan for any change to a live site which deals with transactions no matter how 'small' the site is
The problem these days is the complex interaction between different systems and user traffic means there is no guarantee of a flawless rollout even with a full staging system to test on. I have seen failures occuring due to a unique combination of user requests creating a thread contention which results in an instability causing failure up to 1/2 an hour later!
There is a gap for a system which can capture the entire network traffic being applied to a system and replay it against a staging environment to catch just this sort of problem.
The other problem is a 'minor' problem during rollout such as intermittant failure of some parts. This is let go for a while and people start throwing hacks at the problem. Before you know it enought new transactions have been applied to the system to make rollback unviable.
In my opinion with complex web based systems testing and rollout is becoming one of the most interesting areas in IT at the moment
http://www.upmystreet.com/overview/?l1=N4+1SB#CnvL ink - Links up conversations/messages by Postcode ( UK equiv of Zip Codes but smaller area )
http://www.bbc.co.uk/ican - for folk to organise campaigns one factor is locality
Probably a stupid question this but who's DNA is being mapped here? Is it someone alive? Is it one or many individuals?
Just had a quick look at the tarball. It is a 22 MB substitution phrase based DB read by a reasonably simple PHP script.
It seems to have basis ability to correctly position Proper-Nouns using wild card characters within phrases.
No clever grammar rules etc which is probably a good thing. Stick on a 'did this translate properly' button and let users add to the vocabulary is probably a better approach long term approach with enough users that a clever grammatical algorythm.
Paxman has a huge knowledge of modern politics and politicians which he has used to devastating effect on numerous British politicians in the past
I felt he wasn't as confident or as dangerous with Bill Gates through a lack of technical knowledge.
It is a crying shame than interviewers of the calibre of Paxman rarely have the technical background to deal confidently with these goliaths of IT as they have so succesfully done with politicians.