"If we found out some other retailer is getting a cheaper price on something from the same distributor/manufacturer, our purchasing price will drop below theirs retroactively"
Interestingly I've seen a few things like that here in the UK, but only one company has actually gotten in touch with me after I've bought something to say "Hey our price just dropped - have some money back".
Guess who?
Amazon. They sent me an email to tell me that they'd lowered the prices on something I'd bought literally a day or two before, and then included a discount code for the value of the difference.
That made my day - and would be something realitively easily done by many companys. After all for the large purchases you have your customers names and addresses on file most likely for delivery purposes, right?
D'oh, in between reading the article and writing my comment I'd managed to completely forget about the rsync mention!
curl is "better" than wget since it supports fetching files from servers using SSL, so I'd prefer to use it if I had to pick one or the other.
sudo I use partly for allowing automated tools to perform privileged operations without making them setuid/setgid - as well as for sharing access within temas of admins, and providing an audit trail.
Indeed, the only difference with telnet is that it supports telnet option negotiation - but that's largely irrelevent anyway. As it is only useful for connecting to telnet servers.
For me netcat is so much more useful because it is easily scriptable, for example this fails:
echo -e 'GET/\r\n\r\n' | telnet localhost 80
Replacing telnet with nc works as expected though.
It is suprising the author chose "telnet" as one of the programs in his list.
Sure it is useful for diagnosing random problems, and troubleshooting things - for example connecting straight to a webserver, or simulating a POP3 login request, but I've always preferred netcat.
netcat is much more useful, it allows you to bind to sockets and handling incoming requests as well as make outgoing ones this introduction is a good read.
Missing tools from the list? curl, links/lynx, rsync, sudo, nmap, lsof, and less.
There is one block of adverts on each "article" - on the front page, and the other pages there are none.
(Actually that is not the whole story, there is a 75% chance of viewing an adsense advert upon each article, and a 25% chance of seeing a "paypal donate" button).
I agree that your site is much more open-ended than mine, and has a much wider potential contributor/audience pool.
I'm not unhappy that I'm not raking in $$, just figured that it was worth suggesting that either you got lucky, or are doing more than you suggest, or I'm unlucky.
I've seen other sites where readers contribute most of the content and there is a wide range in the quality of content posted. I guess on/. a good comparision is the number of people who post news links (which are practically no effort) vs the number of people who write soley new content (e.g. book reviews).
I guess that's the situation I'm in, many visitors, many members, few original new pieces of writing. Maybe I just need to reach a critical mass? (And that is certainly not to suggest that I don't value suggestions, or the submissions I've had so far - because I certainly do:)
Choosing the base software was fairly simple, but since then I find I'm making tweaks to the code on an almost daily basis. Sometimes these are just minor things, othertimes I have to make a lot of changes for different reasons.
(Of course switching to a CSS layout to be all cool like/. doesn't help that;)
Even if you allow users to submit content, as I do, there's still a lot of writing I've had to do. With a couple of thousand registered users and a lot more anonymous repeat visitors I still find that only around 1% of users will ever contribute anything.
Most people seem more interested in reading than supplying content - and I find it unlikely this will ever change significantly.
In terms of income I get virtually nothing, personally, the Google Adsense subsidises the site's hosting costs - but doesn't cover it 100%. Still it is a hobby, and it is a useful site for a particular audience so I'll keep it going as long as I can..
He's the one who wins the battle, but he's completely spiritually broken by the experience, can never get over it, and in the end chooses to opt out of the world because he can't keep living with his experiences in a world that pretends nothing's happened.
Not to mention of course that he failed his task.
The fact that the ring was destroyed was due to Gollum. Frodo himself failed (though he did a damn good job up until the moment inside Mount Doom).
I've always thought that was a large part of the reason Frodo was allowed to leave Middle-Earth (that and the wound of the Nazgul.)
There are a lot of ex-z80 users around in the UK as the ZX Spectrum was insanely popular.
I'm just pleased I chose z80 to start with instead of the alternatives such as 6502. As the z80 and x86 instruction sets are very similar it continued to be useful to me as I moved onto PCs from the microcomputers in the 80s. (I think Zilog was helped by an influx of ex-Intel developers, it might be the other way round though!)
Even now I can still remember the opcodes and mappings (e.g. 195 = jmp, 201 = ret) from patching games without the benefit of an assembler or disassembler!
On the other hand I learnt a lot about computers, and had a much simpler time understanding pointers and C concepts than almost anybody else I knew - because I'd grown out of the limitations of BASIC and started writing programs in z80 assembler.
To me the notion that a pointer was merely an address was completely obvious, and I didn't have many problesm with it. Seeing students learning C and struggling is almost painful to me these days.
Because beginning programs nowadays really have it a lot harder than I did. Sure they have pretty GUIs and can do more complex things but they don't need to understand the computer from the bottom up like I did when I was a child.
This invariably means they don't learn the computer, so things like pointers are a mystery to them. (Maybe that is a good thing, maybe not, but it is certainly a different world now). But not knowing what is happening is much like not understanding the code generated by the wizards in Visual Studio. Sooner or later you'll find yourself in over your head and will have no idea what has gone wrong.
(And sure I accept that I had it easy when I started coding in BASIC and had a snappy casette deck to save code on, compared to programmers before me)
I was recently looking for an external hard drive system of some kind to store network backups upon, in addition to centralising my MP3/OGG collection within my home LAN.
One thing that made choosing a device rather difficult was the lack of details on OS supported. Most drives listed 'Windows XP', and I'd assume they were just doing SMB over IP, but it was hard to tell.
Some devices would say things like "The drive appears as a local drive letter on all machines" - at which point I'd think it was some kind of proprietry driver that I'd have little success at using with Linux.
In the end I bought a cheap 160Gb drive which connects to a host using USB2, and now I share that amongst my LAN clients with SMB + NFS.
Not ideal, but certainly better than risking the purchase of something that I couldn't use.
I'll have to investigate more as I'm pretty sure I'll have this drive filled in a matter of a few months.
I reported some DOS bugs against firefox which will kill a browser by essentially saying:
Give me a table of 1000000 rows and 1000000 columns.
The browser dies. Probably because it attempts to either a) allocate all the system's memory and the kernel kills it, or b) at some point memory allocation fails and the program terminates.
Not all crashes are buffer overflows, or exploitable.
Right I'm back from my study - and I have to concurr.
The wires/belts/monitors were annoying, but the nasty bed and constant noise from outside was really distracting, much more so than the monitoring equipment.
As it happens I slept fine for the evening after lying awake for a while.
During the next day I had to sleep for twenty minutes, then get woken up. I think that repeated about four times (something to do with monitoring the early stages of my sleep) and I was barely asleep by the time they came to wake me up - despite normally being able to drop off in a matter of minutes.
Ugh.
Oh well, I'll just hope the results are interesting when they come back and it was worthwhile.
Broken builds seem like they should be simple to detect.
Have a machine download the most recently submitted "ebuild" files, then attempt to build the binaries. Any failures would then result in a new bug being filed automatically.
That would be a useful service to offer - if you wished to help.
Sure you wouldn't catch bugs which were in the binaries, like immediate segfaults, or in configuration file options. But a simple "compile it" test should be trivial to script...
Names are massively non-unique. So claiming a.com for your surname is going to be very hard. Whoever manages it is bound to annoy other people of the same name later. Of the two sites you list the second one seems useful for those who want to be associated with their surname, but didn't think to register the domain themselves - presumably anybody can sign up and be linked with the domain?
If you're interested in surnames specifically you might have a chance at using the.name TLD. That was advertised at one point as being useful for forename.surname.name - not sure how well that worked out because I've rarely seen the domain in the sites I visit.
As for me, I got lucky. I'm all forenamey - and had never thought of using my surname..
Slashdot Mirror
Interestingly I've seen a few things like that here in the UK, but only one company has actually gotten in touch with me after I've bought something to say "Hey our price just dropped - have some money back".
Guess who?
Amazon. They sent me an email to tell me that they'd lowered the prices on something I'd bought literally a day or two before, and then included a discount code for the value of the difference.
That made my day - and would be something realitively easily done by many companys. After all for the large purchases you have your customers names and addresses on file most likely for delivery purposes, right?
Did so :P
b) don't even know what you can do with wget. You can't wget a file via https?I said SSL, not HTTPS - I'm well aware that you can use wget to fetch files over https:/// however you cannot fetch a file over FTP via SSL.
Whilst I accept that might not be a common operation I need to do it often, so curl wins.
D'oh, in between reading the article and writing my comment I'd managed to completely forget about the rsync mention!
curl is "better" than wget since it supports fetching files from servers using SSL, so I'd prefer to use it if I had to pick one or the other.
sudo I use partly for allowing automated tools to perform privileged operations without making them setuid/setgid - as well as for sharing access within temas of admins, and providing an audit trail.
Indeed, the only difference with telnet is that it supports telnet option negotiation - but that's largely irrelevent anyway. As it is only useful for connecting to telnet servers.
For me netcat is so much more useful because it is easily scriptable, for example this fails:
echo -e 'GETReplacing telnet with nc works as expected though.
It is suprising the author chose "telnet" as one of the programs in his list.
Sure it is useful for diagnosing random problems, and troubleshooting things - for example connecting straight to a webserver, or simulating a POP3 login request, but I've always preferred netcat.
netcat is much more useful, it allows you to bind to sockets and handling incoming requests as well as make outgoing ones this introduction is a good read.
Missing tools from the list? curl, links/lynx, rsync, sudo, nmap, lsof, and less.
There is one block of adverts on each "article" - on the front page, and the other pages there are none.
(Actually that is not the whole story, there is a 75% chance of viewing an adsense advert upon each article, and a 25% chance of seeing a "paypal donate" button).
I agree that your site is much more open-ended than mine, and has a much wider potential contributor/audience pool.
I'm not unhappy that I'm not raking in $$, just figured that it was worth suggesting that either you got lucky, or are doing more than you suggest, or I'm unlucky.
I've seen other sites where readers contribute most of the content and there is a wide range in the quality of content posted. I guess on /. a good comparision is the number of people who post news links (which are practically no effort) vs the number of people who write soley new content (e.g. book reviews).
I guess that's the situation I'm in, many visitors, many members, few original new pieces of writing. Maybe I just need to reach a critical mass? (And that is certainly not to suggest that I don't value suggestions, or the submissions I've had so far - because I certainly do :)
I've found that running my site on Debian Administration a fair amount of work.
Choosing the base software was fairly simple, but since then I find I'm making tweaks to the code on an almost daily basis. Sometimes these are just minor things, othertimes I have to make a lot of changes for different reasons.
(Of course switching to a CSS layout to be all cool like /. doesn't help that ;)
Even if you allow users to submit content, as I do, there's still a lot of writing I've had to do. With a couple of thousand registered users and a lot more anonymous repeat visitors I still find that only around 1% of users will ever contribute anything.
Most people seem more interested in reading than supplying content - and I find it unlikely this will ever change significantly.
In terms of income I get virtually nothing, personally, the Google Adsense subsidises the site's hosting costs - but doesn't cover it 100%. Still it is a hobby, and it is a useful site for a particular audience so I'll keep it going as long as I can..
I run a website which shows an advert to unregistered visitors.
Users with an account who are logged in never see them by default, but they do have the option of enabling them if they see fit.
Currently less than 0.5% of members have chosen to enable them.
I was expecting higher; because people have the choice and they are not annoying, but I'm still happy enough, so it all works out.
Not to mention of course that he failed his task.
The fact that the ring was destroyed was due to Gollum. Frodo himself failed (though he did a damn good job up until the moment inside Mount Doom).
I've always thought that was a large part of the reason Frodo was allowed to leave Middle-Earth (that and the wound of the Nazgul.)
There are a lot of ex-z80 users around in the UK as the ZX Spectrum was insanely popular.
I'm just pleased I chose z80 to start with instead of the alternatives such as 6502. As the z80 and x86 instruction sets are very similar it continued to be useful to me as I moved onto PCs from the microcomputers in the 80s. (I think Zilog was helped by an influx of ex-Intel developers, it might be the other way round though!)
Even now I can still remember the opcodes and mappings (e.g. 195 = jmp, 201 = ret) from patching games without the benefit of an assembler or disassembler!
On the other hand I learnt a lot about computers, and had a much simpler time understanding pointers and C concepts than almost anybody else I knew - because I'd grown out of the limitations of BASIC and started writing programs in z80 assembler.
To me the notion that a pointer was merely an address was completely obvious, and I didn't have many problesm with it. Seeing students learning C and struggling is almost painful to me these days.
Because beginning programs nowadays really have it a lot harder than I did. Sure they have pretty GUIs and can do more complex things but they don't need to understand the computer from the bottom up like I did when I was a child.
This invariably means they don't learn the computer, so things like pointers are a mystery to them. (Maybe that is a good thing, maybe not, but it is certainly a different world now). But not knowing what is happening is much like not understanding the code generated by the wizards in Visual Studio. Sooner or later you'll find yourself in over your head and will have no idea what has gone wrong.
(And sure I accept that I had it easy when I started coding in BASIC and had a snappy casette deck to save code on, compared to programmers before me)
Seconded.
I was recently looking for an external hard drive system of some kind to store network backups upon, in addition to centralising my MP3/OGG collection within my home LAN.
One thing that made choosing a device rather difficult was the lack of details on OS supported. Most drives listed 'Windows XP', and I'd assume they were just doing SMB over IP, but it was hard to tell.
Some devices would say things like "The drive appears as a local drive letter on all machines" - at which point I'd think it was some kind of proprietry driver that I'd have little success at using with Linux.
In the end I bought a cheap 160Gb drive which connects to a host using USB2, and now I share that amongst my LAN clients with SMB + NFS.
Not ideal, but certainly better than risking the purchase of something that I couldn't use.
I'll have to investigate more as I'm pretty sure I'll have this drive filled in a matter of a few months.
Yes. All created/tested/reported against 1.04, or earlier.
Still they are examples of the kind of thing you can create if you have enough time upon your hands, and a desire to DOS clients.
I'm pretty sure all are fixed in current versions of Firefox, otherwise I wouldn't have posted the link!
You can get crashes if you use Firefox 1.04, as distributed in Debian Sarge, for example. But that's an older version. (D'oh!)
Not necessarily.
I reported some DOS bugs against firefox which will kill a browser by essentially saying:
The browser dies. Probably because it attempts to either a) allocate all the system's memory and the kernel kills it, or b) at some point memory allocation fails and the program terminates.
Not all crashes are buffer overflows, or exploitable.
The open source site Advogato was hit by a similar profile-page-virus in 2002.
But it was a neat hack, and kudos to Samy!
Right I'm back from my study - and I have to concurr.
The wires/belts/monitors were annoying, but the nasty bed and constant noise from outside was really distracting, much more so than the monitoring equipment.
As it happens I slept fine for the evening after lying awake for a while.
During the next day I had to sleep for twenty minutes, then get woken up. I think that repeated about four times (something to do with monitoring the early stages of my sleep) and I was barely asleep by the time they came to wake me up - despite normally being able to drop off in a matter of minutes.
Ugh.
Oh well, I'll just hope the results are interesting when they come back and it was worthwhile.
Sleep Apnea?
I'm finally going to be tested for that tonight with a full sleep study. I've been waiting for a few months and I literally cannot stay awake.
It appears from talking to partners that I stop breathing fairly regularly when I'm asleep .. and consequently never have a full nights restful sleep.
Shame I was looking forward to the browser updates:
Do you trust this certificate?"A++++ excellent certificate, would do secure business with it again!!! "
Another good site full of first-hand descriptions of how early Apple development was done is http://folklore.org/.
I've never owned a Mac, and am too young to have been involved in earlier developments - but that site does make it all seem very impressive.
Broken builds seem like they should be simple to detect.
Have a machine download the most recently submitted "ebuild" files, then attempt to build the binaries. Any failures would then result in a new bug being filed automatically.
That would be a useful service to offer - if you wished to help.
Sure you wouldn't catch bugs which were in the binaries, like immediate segfaults, or in configuration file options. But a simple "compile it" test should be trivial to script...
I'd love to see you "higher" Gentoo developers. The world needs more tall people ...
And binary packages for Debian's Sarge release on x86 are available here.
Names are massively non-unique. So claiming a .com for your surname is going to be very hard. Whoever manages it is bound to annoy other people of the same name later. Of the two sites you list the second one seems useful for those who want to be associated with their surname, but didn't think to register the domain themselves - presumably anybody can sign up and be linked with the domain?
If you're interested in surnames specifically you might have a chance at using the .name TLD. That was advertised at one point as being useful for forename.surname.name - not sure how well that worked out because I've rarely seen the domain in the sites I visit.
As for me, I got lucky. I'm all forenamey - and had never thought of using my surname..
"DDTT" = "Don't Do That Then".
A fairly common abbreviation I thought ..