That's nice to hear! Too bad nobody cares enough too make web pages that supports it:(
By the way, how does access keys work in IE? Like, in what way is the key to press indicated (like, underlining wouldn't be likely...), and what key combination do you use to activate the link?
Yes, that's it. Too bad iCab is for the Macintosh. But there aren't many pages out there that have the 'accesskey' attributes in place, are there?
Also, you say 'Opera allows you to access nearly all the browser features with keyboard commands' - it was a while (loong while) since I tried out Opera. But doesn't (essentially) all browsers already let me do that? What functions am I forgetting? I think I just use the mouse to click links or select bookmarks (and bookmarks because there are too many of them). Oh, and by the way, a good system for accessing bookmarks via the keyboard would be nice. Something more useful than using the arrows a lot:)
One that does the thing HTML was originally designed to do, but improves on it. Some combination of langauages such as TeX and HTML. Something that describes what the different parts in the page are, and lets the client decide where to place images, if a table of context is needed etc.
AND one that can be used by market-droids to make webpages that the server descides how they sould look. Trying to combine these both two very different goals into one language is obviously not A Good Thing.
THAT is something I really miss about web pages! I want to be able to use browse almost without using the mouse - like, (the most important) links should have keyboard shortcuts. For example, one letter of the link could have a different color - and then you just pressed like ctrl-a-[the link key} and you were en route. This comment field in slashdot should also have a keyboard interface - like 's' to make the 'subject' widget gain focus, 'c' to enter the comment field, and 'a' to check/uncheck the post anonymously checkbox.
And no, don't say 'do it the lynx way'. Tabbing between al links isn't a fast and smooth way. Intelligent keychoices would be.
I think this is supported in (some late) HTML-version - but I doubt there are browsers that support it.
There already are PLENTY of ways in HTTP/1.1 of controling caching behavior. The one you describe, cache images longer than text, is for example easily suported. Also asking the server when a document was last changed is implemented, as well as more advanced stuff like asking for a MD5 sum of the content a given request would return. It's VERY sad very few web sites actually USE these mechanisms. See rfc-2616 and be amazed by how well these things could work!!!!!
Often sites use HTTP/1.1 with persistent connections, and in these cases it stil could be useful. Since almost always all data on one pages comes from one server with one name, info about resolving could stil be useful - and also, info about connecting (or even 'sending SYN',... etc). The info that doesn't fit in the status bar could be displayed in a logwindow of some kind, that you could bring up and look in to see what kind of errors have occured. Here even rendering problems could be told, and stuff like 'your browser doesn't support java, so the applet(s) on this page can't be displayed'.
I agree very much indeed with the original author how wanted verbose loading - I think this is the one thing I miss most in all modern browsers. IE is almost perfect (except the obvious, and that it isn't just a widget that could be fit into other apps (and NO, IWebBrowser isn't good enough since it doesn't let you override enough functions)) except for the lack of descriptive error messages. A very big improvement could be if it told if it was the resolving or the connecting that failed!!
Please... tell me you got the zeros wrong... This stupid being I really hope I'll never meet - tell me he doesn't make 'over $150/hour' for being a lowly receptionist..
If only criminals use encryption, it would be quite easy to find criminals with systems such as your famous carnivore. Unless of course stegenography were to be used.
...but it certainly doesn't matter here if he uses void or int. He could very well know that but be too lazy to write out those boring 'return 0' statements, and so just write void.
and just why would you think he learned C in windows?:)
Equally ironic, I once bought a book about mnemonic techniques. On the back of the book, it told me how much time of my life I would save not having to look for things I've lost.
Needless to say, I lost that book even before I got started reading it.
It may kinda of this topic, but I somewhat like when a new program comes with all eye-candy enabled. That way, I can get quite a good orientation about what the program is capable of offering. Then, I slaughter it to fit my needs. It's a bit lazy, but it's quite convenient. Could of course stab you in the back if you assume that all gui features were shown... Anyway, to somewhat come back on topic, showing of features course shouldn't include anything that constitute a security problem. In this case, I'd argue that the ideal case would have been a requester poping up first time the program is run, quickly informing you about the availability about javascript and its advantages and risks, and asking you if you still wanted it. The REAL question, IMHO, is of course 'why on earth would you want javascript in mail?', but that would be a troll, I guess.
Somebody always mentions this jokey fact yes.. It shouldn't be forgotten that any other mail client would do just fine, even Outlook & co, provided they are sanely configured. PINE just doesn't offer the lame options.
Yeah, problem with MS product is just that you typically need to invert essentially all configuration options before you can start using the program. They obviously come pre-configured for those who don't know how to change configuration.
This demonstrates something common when Linux-people complain about MS.. Spending hours (days!!) setting up X & co is ok, but even going thru the menus in IE5 is 'too much work' for some strange reason. An actual linux-zealot I know complained about the silly 'Go!'-button right of the URL field in IE, and about the fact that IE doesn't do 'autocompletion' of URLs as you type them, and that the smooth scroll is silly, and that those 'friendly error messages' are stupid. All of these are just checkboxes in the configuration. Wierd!
Slashdot Mirror
So, what books did you pick up? Can you make rekommendations about books on the current stand of Genetics?
That's nice to hear! Too bad nobody cares enough too make web pages that supports it :(
By the way, how does access keys work in IE? Like, in what way is the key to press indicated (like, underlining wouldn't be likely...), and what key combination do you use to activate the link?
Yes, that's it. Too bad iCab is for the Macintosh. But there aren't many pages out there that have the 'accesskey' attributes in place, are there? :)
Also, you say 'Opera allows you to access nearly all the browser features with keyboard commands' - it was a while (loong while) since I tried out Opera. But doesn't (essentially) all browsers already let me do that? What functions am I forgetting? I think I just use the mouse to click links or select bookmarks (and bookmarks because there are too many of them). Oh, and by the way, a good system for accessing bookmarks via the keyboard would be nice. Something more useful than using the arrows a lot
I think we need two formats.
One that does the thing HTML was originally designed to do, but improves on it. Some combination of langauages such as TeX and HTML. Something that describes what the different parts in the page are, and lets the client decide where to place images, if a table of context is needed etc.
AND one that can be used by market-droids to make webpages that the server descides how they sould look.
Trying to combine these both two very different goals into one language is obviously not A Good Thing.
How about mozilla and IE? Do they render pages almost identically?
THAT is something I really miss about web pages! I want to be able to use browse almost without using the mouse - like, (the most important) links should have keyboard shortcuts. For example, one letter of the link could have a different color - and then you just pressed like ctrl-a-[the link key} and you were en route. This comment field in slashdot should also have a keyboard interface - like 's' to make the 'subject' widget gain focus, 'c' to enter the comment field, and 'a' to check/uncheck the post anonymously checkbox.
And no, don't say 'do it the lynx way'. Tabbing between al links isn't a fast and smooth way. Intelligent keychoices would be.
I think this is supported in (some late) HTML-version - but I doubt there are browsers that support it.
There already are PLENTY of ways in HTTP/1.1 of controling caching behavior. The one you describe, cache images longer than text, is for example easily suported. Also asking the server when a document was last changed is implemented, as well as more advanced stuff like asking for a MD5 sum of the content a given request would return. It's VERY sad very few web sites actually USE these mechanisms. See rfc-2616 and be amazed by how well these things could work!!!!!
Often sites use HTTP/1.1 with persistent connections, and in these cases it stil could be useful. Since almost always all data on one pages comes from one server with one name, info about resolving could stil be useful - and also, info about connecting (or even 'sending SYN', ... etc). The info that doesn't fit in the status bar could be displayed in a logwindow of some kind, that you could bring up and look in to see what kind of errors have occured. Here even rendering problems could be told, and stuff like 'your browser doesn't support java, so the applet(s) on this page can't be displayed'.
I agree very much indeed with the original author how wanted verbose loading - I think this is the one thing I miss most in all modern browsers. IE is almost perfect (except the obvious, and that it isn't just a widget that could be fit into other apps (and NO, IWebBrowser isn't good enough since it doesn't let you override enough functions)) except for the lack of descriptive error messages. A very big improvement could be if it told if it was the resolving or the connecting that failed!!
Thank you for clearing that up :)
Please... tell me you got the zeros wrong... This stupid being I really hope I'll never meet - tell me he doesn't make 'over $150/hour' for being a lowly receptionist..
If only criminals use encryption, it would be quite easy to find criminals with systems such as your famous carnivore. Unless of course stegenography were to be used.
I'd guess that would be part of the 'mathematical basis' part that was already done.
c'mon, it was VERY obvious that use of encryption wasn't what was relevant in this article
typing wrong ip/domain name has equally much to do with being a CS major as do remembering phone numbers or not making typos during exams...
...but it certainly doesn't matter here if he uses void or int. He could very well know that but be too lazy to write out those boring 'return 0' statements, and so just write void.
:)
and just why would you think he learned C in windows?
yeah, this is absolutely nothing new. Even dr. Watson had a notebook.
Equally ironic, I once bought a book about mnemonic techniques. On the back of the book, it told me how much time of my life I would save not having to look for things I've lost.
Needless to say, I lost that book even before I got started reading it.
It may kinda of this topic, but I somewhat like when a new program comes with all eye-candy enabled. That way, I can get quite a good orientation about what the program is capable of offering. Then, I slaughter it to fit my needs. It's a bit lazy, but it's quite convenient. Could of course stab you in the back if you assume that all gui features were shown...
Anyway, to somewhat come back on topic, showing of features course shouldn't include anything that constitute a security problem. In this case, I'd argue that the ideal case would have been a requester poping up first time the program is run, quickly informing you about the availability about javascript and its advantages and risks, and asking you if you still wanted it.
The REAL question, IMHO, is of course 'why on earth would you want javascript in mail?', but that would be a troll, I guess.
Somebody always mentions this jokey fact yes.. It shouldn't be forgotten that any other mail client would do just fine, even Outlook & co, provided they are sanely configured. PINE just doesn't offer the lame options.
Yeah, problem with MS product is just that you typically need to invert essentially all configuration options before you can start using the program. They obviously come pre-configured for those who don't know how to change configuration.
This demonstrates something common when Linux-people complain about MS.. Spending hours (days!!) setting up X & co is ok, but even going thru the menus in IE5 is 'too much work' for some strange reason. An actual linux-zealot I know complained about the silly 'Go!'-button right of the URL field in IE, and about the fact that IE doesn't do 'autocompletion' of URLs as you type them, and that the smooth scroll is silly, and that those 'friendly error messages' are stupid. All of these are just checkboxes in the configuration. Wierd!