I'd like to have a lexer for Scintilla for (basically) the TWiki markup language. It can be written in C++ (preferred) or Lua, or possibly some other ways (as an "external" lexer).
I'm looking into it, but I'm not a C nor C++ (nor a Lua) programmer--if you're interested in helping, give me a shout.
Re: "it is designed very, very carefully to allow documents to be rendered differently"
Can you shed some light on why they would have done that (short of dealing with browsers with lesser capabilities, the biggest example being text only browsers)?
I think it finally got said somewhere down below, but the Sugar website points out that you can create a boot CD-Rom and use that to boot to the USB drive.
I suppose you might be able to run from the USB drive image on a CD-Rom, but they explicitly point out (the obvious) that you won't be able to save your work to the CD-Rom.
Re: "Microsoft used to release patches as soon as they were discovered. They worked that way for decades. A hole was found, a fix was built, tested, and released."
That's not quite what I understood--it might be correct if rephrased: "Microsoft used to release patches as soon as a patch was developed" (and, I hope, tested).
I didn't pay careful attention to relevant reports to confirm that it was often or usually or even ever Microsoft, but I understood it was quite common to sit on a known vulnerability for quite a long time before releasing a patch.
Re: "A classic salestard in action, and yet another reason to shop online."
Maybe, but I wouldn't be surprised to find out that is the (retailing) company's intended sales pitch--there is surely more margin in newer (in this case wireless) stuff than older (wired) stuff.
I usually try to bottom post on emails, but I think something should be recognized, bottom posting is a new approach, and top posting is the old tried and true approach (at least generally).
Millions (??) of business writers have been trained to encapsulate the most important part of a business letter in the first paragraph. Most important things (and a summary) at the top, less important things (and details) later.
Similarly, tens of thousands (??) of journalists have been trained in a similar fashion--most important stuff in the first paragraph, less important stuff and details later in the article (and, in fact, ranked (based on somebody's judgment) of order of importance. Among other things, so that if an editor has to shorten an article, he just cuts stuff off the end. (Likewise, a reader can read just the beginning of an article to get the basic gist.)
Also, similarly, for a business reader, she can get the gist from the first paragraph.
So now email comes along and (nominally) espouses a different approach (yes, I know email can be structured with bottom posting to cover the most important points first, but I'm not sure how much thought is given to that)--what do you (the collective you) expect to happen?
With respect to email, I find my own habits changing and am starting to do more top posting. I have reordered quoted parts of an email I'm to which I'm responding in order to have a sequence more conducive to the point(s) I want to make.
In retrospect, in the last line, I should have included the word "different" (in parenthesis), i.e.:
I guess this is particularly ironic (I think that's the right word)--to comment on the use of a phrase (or word) in a comment that is already about the use of a (different) word.;-)
I'm somewhat embarrassed to make this comment here (partly because this Slashdot thread has some very good discussion on intellectual property), but I've seen this signature (or something like it) a few times and wanted to respond somewhere.
Re your signature: 'For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares?"'
Did you really mean "intensive purposes"? The original phrase, to the best of my knowledge, was "for all intents and purposes". Not to say you (or someone) might really mean "intensive purposes", but then I'd ask you what you mean by that.
I guess this is particularly ironic (I think that's the right word)--to comment on the use of a phrase (or word) in a comment that is already about the use of a word.;-)
Sorry, I haven't read all the comments here--some of the following are based on other comments, some are not (or not based on comments I read).
* First, I'd start with estimating the value of the data, or the cost of losing the data, or having a malicious 3rd party obtain the data.
* If that is a great deal, I'd consider not letting a contractor work on the data offsite, but instead insist on one of the following:
* The contractor work on site, under security procedures established specifically for this work
* If you need a contractor to develop the procedures to import the data to the new package, give her a sample (invalid) dataset and let her develop the procedures on that. She should provide a set of tests to confirm the procedures work properly, but you should have your own additional testing to confirm that the sample dataset hasn't missed any corner cases.
* Or, you may need the contractor to teach an "inside" employee how to import the data, with the contractor at "arms length" from the data. (Of course, this begs the question (in the UK sense) of how to make sure the inside employee (and your inside data security procedures) are adequate.
* I don't know how well bonding works, but if I did let the data offsite, I'd have the contractor post a gigantic bond, appropriate to the cost of data loss. (An issue here is how long the bond stays in effect--ideally it should be until the data loses it's value due to "attrition" (of whatever sort) over time.)
Another miscellaneous note--the idea that decryption takes a long time is based on averages (or whatever)--there is a chance that an attacker will guess the key on an early try. If I was an attacker, and the data was valuable enough, I might set an automated cracker to work on it on the off chance that I might get lucky.
Re:Would a caching proxy (like Squid) help?
on
Dealing With Dialup
·
· Score: 1
Oops, never mind--I missed the other comment that talked about exactly this. (ebkac)
Would a caching proxy (like Squid) help?
on
Dealing With Dialup
·
· Score: 1
I never got around to setting it up, but I often wondered if a caching proxy (like Squid) would help with web browsing, especially in situations where you often visit pages from the same site (like reading slashdot).
Anybody have any experience?
Maybe you'd have to do something to fool some of the timeouts (can that be done)--I mean the things that say a page (or image, or whatever) has expired?
I recommend that you try (or retry) black (or dark) text on a white (or light) background, but *turn the brightness down".
I find so many monitor with the brightness turned all the way up, which has to be hard on the monitor itself (the screen and internal electronics), has to be emitting more radiation, and is probably using more power.
the system is not first to file but first to invent
I guess that is still true today, but isn't legislation (or something) pending to change our patent system to first to file?
And even if it isn't pending at the moment, or has recently been defeated, many other countries' patent systems are first to file, and that seems to be the trend. So, even if that change has been defeated recently, I'm sure it will come up again.
My point: the advice in this article should account for that (or at least mention it), with strategies that work for first to invent as well as first to file, or, if that is not possible, discuss strategies for both situations, or something.
I guess I'm not sure what a consumable is (or how it gets treated), but when I was heavily involved in business, treating something as an asset to be depreciated was a disadvantage, not an advantage--an asset had to be written off (against income) a little bit each year (varying depending on the class of asset), whereas something that could be expensed could be written off (against income) in the current year.
Things that were consumable (e.g., roof bolts) in our business (coal mining and steel making) were written off against expenses in the current year.
Let's say I go out and buy a bunch of supplies (compilers and debuggers) and go to school to learn a bunch of architectural skills (programming knowledge). Then, using my new found resources, I build a house (program). It's my damn decision to live in it by myself, burn it down, rent it out, leave it empty, or give it away. What he's saying is that other people breaking into my house and stealing all or parts out of it at their leisure is okay as long as they're willing to share their loot with the rest of the world.
Let's start from a few different places:
* In general, nobody is forcing you to apply the GPL to your code, except in one case (below)
* In some cases (more below), if some of the supplies you "bought" were GPL'd, the price of those supplies includes that what you create from those supplies also be licensed under the GPL.
To try to clarify the ambiguous statement I just made, the use of GPL'd compilers and debuggers (i.e., tools) does not require that you license what you produce using those under the GPL.
On the other hand, if you modify an existing GPL'd program, your building supplies (as opposed to tools) were GPL'd, and you are required to license your modifications under the GPL.
Just as you can choose to license something you make for dollars, others can choose to license something they make for other forms of renumeration (sp?). One form of renumeration is a guarantee that the work they performed will always be available to others, to use, study,... (the four freedoms that Richard Stallman points to).
Again, if you have the right to license your program for dollars, I have the right to license mine for dollars or something else that I value. Depending on what building supplies (as opposed to tools) you choose to use (public domain, proprietary, or GPL'd), you have to pay the cost of those. Some are free, some will cost you money, and some may cost you something else.
Oops, in that last scenario for doze I should have deleted the left clicks on steps 3 and 5, so it should read:
Doze:
1. [Mouse] Left-click and select text to be pasted. 2. [Keyboard] CTRL-C 3. [Mouse] Select (an instance of) text to be replaced 4. [Keyboard] CTRL-V 5. [Mouse] Select next instance of text to be replaced 6. [Keyboard] CTRL-V 7. Repeat 5 and 6 as often as desired
Ok, but consider these three variations. (I think there might be workarounds in the Unix approach, but I don't know what they are):
Cut and Paste:
X:
1. [Mouse] Left-Click and select Text. 2. [Mouse] Left-Click destination. 3. [Mouse] Middle-Button paste. 4. [Keyboard] Press Delete 5. Oops, your cursor is back where you copied the text from, not where you copied the text to (which is not what I ususally want)
Cut or Copy and Paste Multiple Times, Scrolling the Screen to Get to Subsequent Paste Locations
X:
1. [Mouse] Left-Click and select Text. 2. [Mouse] Left-Click destination. 3. [Mouse] Middle-Button paste. 4. Scroll (however) -- oops, you've (probably) lost the selection 5. ?? Start over??
Doze:
1. [Mouse] Left-click and select text. 2. [Keyboard] CTRL-X 3. [Mouse] Left-click destination. 4. [Keyboard] CTRL-V 5. Scroll (however) 6. [Keyboard] CTRL-V 7. Repeat 5 and 6 as often as desired
Copy and Paste in place of Something Else (Multiple times if you are a glutton for punishment)
X:
1. Go to (all) location(s) of text to be replaced, select and delete it (and remember those locations) 2. Go to location of text to be pasted in, [Mouse] Left-Click and select Text. 3. [Mouse] Left-Click destination. 4. [Mouse] Middle-Button paste.
Yes, you could do a search and replace, and sometimes I do, it depends on what seems most convenient at the moment.
Doze:
1. [Mouse] Left-click and select text to be pasted. 2. [Keyboard] CTRL-C 3. [Mouse] Select (an instance of) text to be replaced -- Left-click 4. [Keyboard] CTRL-V 5. [Mouse] Select next instance of text to be replaced -- Left-click 6. [Keyboard] CTRL-V 7. Repeat 5 and 6 as often as desired
BTW: I like klipper quite a bit, but it would be nicer if the buffers in klipper each had a shortcut paste key, so I could copy multiple items and paste them in multiple locations by pressing, for example, CTRL-V1, -V2,...
Yes, I'd have to remember what is in each buffer, but I can usually manage that for a short time, or view the klipper buffers again to see what is where.
This also depends on not having "accidental" additions to the klipper buffers because you've selected text for some operation other than copy and paste. You can set klipper to not sync the selection to the clipboard to minimize that, although it still seems to happen -- oops, maybe I have to also "ignore the selection" -- I'll try that.
Re:I wouldn't call it "goo"
on
Universal Goo
·
· Score: 1
Sorry, this is definitely OT. It would be nice (AFAIC) if long URLs were incorporated in the same link
Hmm, that [reflectionsoldiers.com] thingie is kind of neat and it's automatic.
To accomplish that, use the following syntax and post as "HTML Formatted". Unfortunately, one problem with posting as HTML Formatted is that you must then use <p \> to create blank lines to separate paragraphs.
<a href="http://www.reflectionsoldiers.com/forumtest/ viewforum.php?f=2&sid=7ea89aa14629180fff24300e7e87 4225">the same link</a>
Are there alternate approaches? (Especially to the need to post as HTML Formatted?)
I'd like to get the word out. If you are so inclined, please spread the word. If you have suggestions on how to spread the word (or anything else), make your comments and suggestions this WikiLearn page.
> 9600? Hmmmm... WHY!?! Last I checked 56k modems ran for about $5 a piece (less in bulk).
Well, I live too far from my telephone central office for DSL, my cable company offers only one way modems (in my neigborhood), the best my 56k modem can connect at is around 33k, and I share the connection with two other members of my family.
Something that operates on a 9.6k line would be nice!
Slashdot Mirror
I'd like to have a lexer for Scintilla for (basically) the TWiki markup language. It can be written in C++ (preferred) or Lua, or possibly some other ways (as an "external" lexer).
I'm looking into it, but I'm not a C nor C++ (nor a Lua) programmer--if you're interested in helping, give me a shout.
Thanks!
Re: "it is designed very, very carefully to allow documents to be rendered differently"
Can you shed some light on why they would have done that (short of dealing with browsers with lesser capabilities, the biggest example being text only browsers)?
I think it finally got said somewhere down below, but the Sugar website points out that you can create a boot CD-Rom and use that to boot to the USB drive.
I suppose you might be able to run from the USB drive image on a CD-Rom, but they explicitly point out (the obvious) that you won't be able to save your work to the CD-Rom.
Re: "Microsoft used to release patches as soon as they were discovered. They worked that way for decades. A hole was found, a fix was built, tested, and released."
That's not quite what I understood--it might be correct if rephrased: "Microsoft used to release patches as soon as a patch was developed" (and, I hope, tested).
I didn't pay careful attention to relevant reports to confirm that it was often or usually or even ever Microsoft, but I understood it was quite common to sit on a known vulnerability for quite a long time before releasing a patch.
Re: "A classic salestard in action, and yet another reason to shop online."
Maybe, but I wouldn't be surprised to find out that is the (retailing) company's intended sales pitch--there is surely more margin in newer (in this case wireless) stuff than older (wired) stuff.
I usually try to bottom post on emails, but I think something should be recognized, bottom posting is a new approach, and top posting is the old tried and true approach (at least generally).
Millions (??) of business writers have been trained to encapsulate the most important part of a business letter in the first paragraph. Most important things (and a summary) at the top, less important things (and details) later.
Similarly, tens of thousands (??) of journalists have been trained in a similar fashion--most important stuff in the first paragraph, less important stuff and details later in the article (and, in fact, ranked (based on somebody's judgment) of order of importance. Among other things, so that if an editor has to shorten an article, he just cuts stuff off the end. (Likewise, a reader can read just the beginning of an article to get the basic gist.)
Also, similarly, for a business reader, she can get the gist from the first paragraph.
So now email comes along and (nominally) espouses a different approach (yes, I know email can be structured with bottom posting to cover the most important points first, but I'm not sure how much thought is given to that)--what do you (the collective you) expect to happen?
With respect to email, I find my own habits changing and am starting to do more top posting. I have reordered quoted parts of an email I'm to which I'm responding in order to have a sequence more conducive to the point(s) I want to make.
In retrospect, in the last line, I should have included the word "different" (in parenthesis), i.e.:
;-)
I guess this is particularly ironic (I think that's the right word)--to comment on the use of a phrase (or word) in a comment that is already about the use of a (different) word.
I'm somewhat embarrassed to make this comment here (partly because this Slashdot thread has some very good discussion on intellectual property), but I've seen this signature (or something like it) a few times and wanted to respond somewhere.
;-)
Re your signature: 'For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares?"'
Did you really mean "intensive purposes"? The original phrase, to the best of my knowledge, was "for all intents and purposes". Not to say you (or someone) might really mean "intensive purposes", but then I'd ask you what you mean by that.
I guess this is particularly ironic (I think that's the right word)--to comment on the use of a phrase (or word) in a comment that is already about the use of a word.
Sorry, I haven't read all the comments here--some of the following are based on other comments, some are not (or not based on comments I read).
* First, I'd start with estimating the value of the data, or the cost of losing the data, or having a malicious 3rd party obtain the data.
* If that is a great deal, I'd consider not letting a contractor work on the data offsite, but instead insist on one of the following:
* The contractor work on site, under security procedures established specifically for this work
* If you need a contractor to develop the procedures to import the data to the new package, give her a sample (invalid) dataset and let her develop the procedures on that. She should provide a set of tests to confirm the procedures work properly, but you should have your own additional testing to confirm that the sample dataset hasn't missed any corner cases.
* Or, you may need the contractor to teach an "inside" employee how to import the data, with the contractor at "arms length" from the data. (Of course, this begs the question (in the UK sense) of how to make sure the inside employee (and your inside data security procedures) are adequate.
* I don't know how well bonding works, but if I did let the data offsite, I'd have the contractor post a gigantic bond, appropriate to the cost of data loss. (An issue here is how long the bond stays in effect--ideally it should be until the data loses it's value due to "attrition" (of whatever sort) over time.)
Another miscellaneous note--the idea that decryption takes a long time is based on averages (or whatever)--there is a chance that an attacker will guess the key on an early try. If I was an attacker, and the data was valuable enough, I might set an automated cracker to work on it on the off chance that I might get lucky.
Oops, never mind--I missed the other comment that talked about exactly this. (ebkac)
I never got around to setting it up, but I often wondered if a caching proxy (like Squid) would help with web browsing, especially in situations where you often visit pages from the same site (like reading slashdot).
Anybody have any experience?
Maybe you'd have to do something to fool some of the timeouts (can that be done)--I mean the things that say a page (or image, or whatever) has expired?
Randy Kramer
Is this going to be sort of a poison pill?
If anybody reads the Microsoft documents might they be subject to triple penalties for use of patented material?
Randy Kramer
I recommend that you try (or retry) black (or dark) text on a white (or light) background, but *turn the brightness down".
I find so many monitor with the brightness turned all the way up, which has to be hard on the monitor itself (the screen and internal electronics), has to be emitting more radiation, and is probably using more power.
Randy Kramer
I guess that is still true today, but isn't legislation (or something) pending to change our patent system to first to file?
And even if it isn't pending at the moment, or has recently been defeated, many other countries' patent systems are first to file, and that seems to be the trend. So, even if that change has been defeated recently, I'm sure it will come up again.
My point: the advice in this article should account for that (or at least mention it), with strategies that work for first to invent as well as first to file, or, if that is not possible, discuss strategies for both situations, or something.
Re: Treating PCs as assets, not consumables:
I guess I'm not sure what a consumable is (or how it gets treated), but when I was heavily involved in business, treating something as an asset to be depreciated was a disadvantage, not an advantage--an asset had to be written off (against income) a little bit each year (varying depending on the class of asset), whereas something that could be expensed could be written off (against income) in the current year.
Things that were consumable (e.g., roof bolts) in our business (coal mining and steel making) were written off against expenses in the current year.
Except maybe the elevator cleaning person?
Let's start from a few different places:
* In general, nobody is forcing you to apply the GPL to your code, except in one case (below)
* In some cases (more below), if some of the supplies you "bought" were GPL'd, the price of those supplies includes that what you create from those supplies also be licensed under the GPL.
To try to clarify the ambiguous statement I just made, the use of GPL'd compilers and debuggers (i.e., tools) does not require that you license what you produce using those under the GPL.
On the other hand, if you modify an existing GPL'd program, your building supplies (as opposed to tools) were GPL'd, and you are required to license your modifications under the GPL.
Just as you can choose to license something you make for dollars, others can choose to license something they make for other forms of renumeration (sp?). One form of renumeration is a guarantee that the work they performed will always be available to others, to use, study,
Again, if you have the right to license your program for dollars, I have the right to license mine for dollars or something else that I value. Depending on what building supplies (as opposed to tools) you choose to use (public domain, proprietary, or GPL'd), you have to pay the cost of those. Some are free, some will cost you money, and some may cost you something else.
Oops, in that last scenario for doze I should have deleted the left clicks on steps 3 and 5, so it should read:
Doze:
1. [Mouse] Left-click and select text to be pasted.
2. [Keyboard] CTRL-C
3. [Mouse] Select (an instance of) text to be replaced
4. [Keyboard] CTRL-V
5. [Mouse] Select next instance of text to be replaced
6. [Keyboard] CTRL-V
7. Repeat 5 and 6 as often as desired
Ok, but consider these three variations. (I think there might be workarounds in the Unix approach, but I don't know what they are):
...
Cut and Paste:
X:
1. [Mouse] Left-Click and select Text.
2. [Mouse] Left-Click destination.
3. [Mouse] Middle-Button paste.
4. [Keyboard] Press Delete
5. Oops, your cursor is back where you copied the text from, not where you copied the text to (which is not what I ususally want)
Doze:
1. [Mouse] Left-click and select text.
2. [Keyboard] CTRL-X
3. [Mouse] Left-click destination.
4. [Keyboard] CTRL-V
Cut or Copy and Paste Multiple Times, Scrolling the Screen to Get to Subsequent Paste Locations
X:
1. [Mouse] Left-Click and select Text.
2. [Mouse] Left-Click destination.
3. [Mouse] Middle-Button paste.
4. Scroll (however) -- oops, you've (probably) lost the selection
5. ?? Start over??
Doze:
1. [Mouse] Left-click and select text.
2. [Keyboard] CTRL-X
3. [Mouse] Left-click destination.
4. [Keyboard] CTRL-V
5. Scroll (however)
6. [Keyboard] CTRL-V
7. Repeat 5 and 6 as often as desired
Copy and Paste in place of Something Else (Multiple times if you are a glutton for punishment)
X:
1. Go to (all) location(s) of text to be replaced, select and delete it (and remember those locations)
2. Go to location of text to be pasted in, [Mouse] Left-Click and select Text.
3. [Mouse] Left-Click destination.
4. [Mouse] Middle-Button paste.
Yes, you could do a search and replace, and sometimes I do, it depends on what seems most convenient at the moment.
Doze:
1. [Mouse] Left-click and select text to be pasted.
2. [Keyboard] CTRL-C
3. [Mouse] Select (an instance of) text to be replaced -- Left-click
4. [Keyboard] CTRL-V
5. [Mouse] Select next instance of text to be replaced -- Left-click
6. [Keyboard] CTRL-V
7. Repeat 5 and 6 as often as desired
BTW: I like klipper quite a bit, but it would be nicer if the buffers in klipper each had a shortcut paste key, so I could copy multiple items and paste them in multiple locations by pressing, for example, CTRL-V1, -V2,
Yes, I'd have to remember what is in each buffer, but I can usually manage that for a short time, or view the klipper buffers again to see what is where.
This also depends on not having "accidental" additions to the klipper buffers because you've selected text for some operation other than copy and paste. You can set klipper to not sync the selection to the clipboard to minimize that, although it still seems to happen -- oops, maybe I have to also "ignore the selection" -- I'll try that.
Hmm, that [reflectionsoldiers.com] thingie is kind of neat and it's automatic.
To accomplish that, use the following syntax and post as "HTML Formatted". Unfortunately, one problem with posting as HTML Formatted is that you must then use <p \> to create blank lines to separate paragraphs.
<a href="http://www.reflectionsoldiers.com/forumtest/ viewforum.php?f=2&sid=7ea89aa14629180fff24300e7e87 4225">the same link</a>
Are there alternate approaches? (Especially to the need to post as HTML Formatted?)
I'd like to get the word out. If you are so inclined, please spread the word. If you have suggestions on how to spread the word (or anything else), make your comments and suggestions this WikiLearn page.
Don't forget flubber (or is that a compound?)
> 9600? Hmmmm... WHY!?! Last I checked 56k modems ran for about $5 a piece (less in bulk).
Well, I live too far from my telephone central office for DSL, my cable company offers only one way modems (in my neigborhood), the best my 56k modem can connect at is around 33k, and I share the connection with two other members of my family.
Something that operates on a 9.6k line would be nice!
Pinpoint fire == lasers!
I wonder if I could patent this now, or rather if this qualifies as "prior art" to keep somebody else from patenting this obvious idea.
It will probably be quite a few years it will need a very accurate aiming system and stable platform.
Scientist: research, investigation
Engineer: design
Technician: implementation