Why is Everybody so clueless on the importance of Desktop Search to the Masses?
-
Wow, I just stumbled on some more forum postings by SpectateSwamp over on a nature photography forum, where he posts video of dying/trapped animals (unedited I assume, so he can use SpectateSearch to index it and randomly get "re aquanted" with each dead animal). At best this guy is a dick and at worse he's a perverse loser with too much free time. Someone find him a consulting gig or something.
-
@MasterPlanSoftware said:
@dlikhten said:
@MasterPlanSoftware said:
@tster said:
I wonder how long it would take to write a search that's faster than his. I wouldn't think it would take longer than a couple minutes in Perl...
Quick question:
Does DesktopSearch support searching with regular expressions?
No. It is no more advanced than notepad's find function. And I cannot even say it is THAT good.
<rant>
Just 1 question, seeing as how GREP is under GNU could spotlight at least have just taken GREP from UNIX UTILS (windows binaries) and just incorporated it inside it's run time. It would only have to mention that it uses the GNU grep. HENCE OPEN SOURCE. My god people seem to LOVE re-inventing wheel, and putting pattens on them. Sigh
</rant>
I knew it! You ARE SpectateSwamp!
Who? I just picked up the tags from this forum :)
I say we start the <ambiguously troll> posting tags. Are you trolling? Are you not? Nobody knows! What's everyone staring at?
-
@dlikhten said:
@MasterPlanSoftware said:
@dlikhten said:
@MasterPlanSoftware said:
@tster said:
I wonder how long it would take to write a search that's faster than his. I wouldn't think it would take longer than a couple minutes in Perl...
Quick question:
Does DesktopSearch support searching with regular expressions?
No. It is no more advanced than notepad's find function. And I cannot even say it is THAT good.
<rant>
Just 1 question, seeing as how GREP is under GNU could spotlight at least have just taken GREP from UNIX UTILS (windows binaries) and just incorporated it inside it's run time. It would only have to mention that it uses the GNU grep. HENCE OPEN SOURCE. My god people seem to LOVE re-inventing wheel, and putting pattens on them. Sigh
</rant>
I knew it! You ARE SpectateSwamp!
Who? I just picked up the tags from this forum :)
I say we start the <ambiguously troll> posting tags. Are you trolling? Are you not? Nobody knows! What's everyone staring at?
Though replying to tster's original post... right on. Searching + perl = perlarific.
-
@dlikhten said:
@MasterPlanSoftware said:
@dlikhten said:
@MasterPlanSoftware said:
@tster said:
I wonder how long it would take to write a search that's faster than his. I wouldn't think it would take longer than a couple minutes in Perl...
Quick question:
Does DesktopSearch support searching with regular expressions?
No. It is no more advanced than notepad's find function. And I cannot even say it is THAT good.
<rant>
Just 1 question, seeing as how GREP is under GNU could spotlight at least have just taken GREP from UNIX UTILS (windows binaries) and just incorporated it inside it's run time. It would only have to mention that it uses the GNU grep. HENCE OPEN SOURCE. My god people seem to LOVE re-inventing wheel, and putting pattens on them. Sigh
</rant>
I knew it! You ARE SpectateSwamp!
Who? I just picked up the tags from this forum :)
I say we start the <ambiguously troll> posting tags. Are you trolling? Are you not? Nobody knows! What's everyone staring at?
Not trolling as much, since I don't care about a reaction... but your 'ideas' and 'rants' bear a resemblance to our Swamp friend.
-
Has anyone made commentary on how SS uses Telus? Everyone raves about it (being shitty), it really seems like another angle to poke him.
@everyone said:
*pokes Spectateswamp*
HAHA, you don't even know a good ISP from a shit one
EDIT: shit, forgot to feed the troll again
-
Telus has been crap lately
@belgariontheking said:
Has anyone made commentary on how SS uses Telus? Everyone raves about it (being shitty), it really seems like another angle to poke him.
@everyone said:
*pokes Spectateswamp*
HAHA, you don't even know a good ISP from a shit one
EDIT: shit, forgot to feed the troll again
Not a troll. A desktop search advocate maybe. Experience tells me that I can teach someone how to use this program in an afternoon. They will understand what is going on. There is no big mystery. In this day and age you can do things inefficiently. Computers are so fast it doesn't matter. I programmed in the 70's and know what crap talking about performance today is. This is not rocket science. You have the code. Go do something constructive. Make it better.
-
@SpectateSwamp said:
@belgariontheking said:
Has anyone made commentary on how SS uses Telus? Everyone raves about it (being shitty), it really seems like another angle to poke him.
@everyone said:
*pokes Spectateswamp*
HAHA, you don't even know a good ISP from a shit one
EDIT: shit, forgot to feed the troll again
Not a troll. A desktop search advocate maybe. Experience tells me that I can teach someone how to use this program in an afternoon. They will understand what is going on. There is no big mystery. In this day and age you can do things inefficiently. Computers are so fast it doesn't matter. I programmed in the 70's and know what crap talking about performance today is. This is not rocket science. You have the code. Go do something constructive. Make it better.
You definitely are a troll. If you cannot install a program, and have it just work without reading something, watching a video or visiting some creepy freak's shack, then it is not worth of being in the public domain.
Also, don't call your VB6 app a desktop search. It is not. It doesn't even pretend to try to be. It doesn't even live up to Notepad's find function: http://www.mediafire.com/?bc7zuy2s0lc
No one will ever work on this piece of crap, try all you want, it will never happen.
-
No one really thinks he's a bot, do they? If he's a bot, then the author not only wrote a huge VB project, he also wrote the best bot I've ever seen (grammar not withstanding), but decided to use it only for occasional semi-trolling.
-
@Cap'n Steve said:
No one really thinks he's a bot, do they? If he's a bot, then the author not only wrote a huge VB project, he also wrote the best bot I've ever seen (grammar not withstanding), but decided to use it only for occasional semi-trolling.
I had a suspicion early on, but only based on a few posts. It is becoming pretty obvious he is just a nutcase who doesn't understand the rest of the world that passed him by.
asuffield is the one pushing the bot theory. I would love to hear someone describe this algorithm... it must have been compiled in a mental institution under a full moon.
-
Not so ClueLess any more
Bot or not. I'm happy. a number of you have taken and tried the search. I don't blame you for resisting the search. I used search more than anybody I know. All the time for the last 10 or 15 years. It's simple enough that anybody can quickly pick up what's happening. When you have the code. You might get more interested in checking it out. The Search doesn't do anything unusual. I can fill you in on more of the details. Maybe when the laughter stops. And sombody asks
-
-
@SpectateSwamp said:
The Search doesn't do anything unusual.
I would argue that displaying a random image and/or video, requiring you to create your own index, and requiring you to export your emails to text is a strange thing for a search to do.
@SpectateSwamp said:
I can fill you in on more of the details. Maybe when the laughter stops. And sombody asks
Well neither are going to happen from me.
-
@belgariontheking said:
You're the one spending all day on forum boards spreading your seed of bad programs. You know activities, don't you?
I, for one, did not need that particular mental image.
-
SpectateSwamp:
Some questions for you and your offering:
1. Can it search an entire hard disk of files? (files of which cannot be concatenated into one large file - such as legal documentation, files of different formats)
2. (Borrowing from someone else): Can it search using regular expressions? (Perl-compatible is better)
3. Since your tool doesn't have an index, can you give us some benchmarks on search speeds, CPU utilization and execution time on different types of searches?
4. Can it search through PDFs, PostScript, RTF documents and other formatted files? (I ask this as I wish to know if your tool can search though manuals in formats such as PDFs and other formatted documents that cannot be transformed into plaintext files such as PCFs which commonly have graphics).
Unfortunately, I don't have a Windows system with me to test your tool so I am unable to use it personally. (I use linux)
-
@DigitalXeron said:
SpectateSwamp:
Some questions for you and your offering:
1. Can it search an entire hard disk of files? (files of which cannot be concatenated into one large file - such as legal documentation, files of different formats)
2. (Borrowing from someone else): Can it search using regular expressions? (Perl-compatible is better)
3. Since your tool doesn't have an index, can you give us some benchmarks on search speeds, CPU utilization and execution time on different types of searches?
4. Can it search through PDFs, PostScript, RTF documents and other formatted files? (I ask this as I wish to know if your tool can search though manuals in formats such as PDFs and other formatted documents that cannot be transformed into plaintext files such as PCFs which commonly have graphics).
Unfortunately, I don't have a Windows system with me to test your tool so I am unable to use it personally. (I use linux)
His replies will (sadly) be:
1. Blah Blah Blah you should combine all your files on your computer into one giant text file.
2. What are regular expressions?
3. It is very fast. But the code sucks. But who needs optimization? Vax. Something about the 80's.
4. You can export all of these files into text, and then they can be searched.
If he answers, that is what it will amount to. None of it will make any sense at all, and you will be left with a headache.
-
@DigitalXeron said:
fortunately, I don't have a Windows system with me to test your tool so I am unable to use it personally. (I use linux)
fixed that for you.
-
@DigitalXeron said:
1. Can it search an entire hard disk of files? (files of which cannot be concatenated into one large file - such as legal documentation, files of different formats)
No, it can't, because it's not a "search" app. It does not scan a drive, it does not parse media files, all it can do is open a text file and do a very braindead static word search. It's basically notepad that has some minor abilities to open random files (this is apparently one of Swamp's major selling points - randomly opening files). If what you have cannot be put into the single solitary text file that SSDS can look through, it won't ever be "found".
@DigitalXeron said:
2. (Borrowing from someone else): Can it search using regular expressions? (Perl-compatible is better)
No. That's geek stuff and he doesn't/can't/won't/refuses to do anything that might be actually useful. Essentially all he does is slurp up a text file and see if whatever VB's equivalent of C's strstr() returns true.
@DigitalXeron said:
3. Since your tool doesn't have an index, can you give us some benchmarks on search speeds, CPU utilization and execution time on different types of searches?
Ah, but it does have an index. You have to sit there and manually type up indexing data on all your videos, images, binary-only files (.doc/pdf/ps/etc....). It won't build that index for you, because it doesn't "scan" anything except this index file you build yourself. You don't even need SSDS for building it either, you might as well use Notepad and get a real (well, better) editor.
@DigitalXeron said:
4. Can it search through PDFs, PostScript, RTF documents and other formatted files? (I ask this as I wish to know if your tool can search though manuals in formats such as PDFs and other formatted documents that cannot be transformed into plaintext files such as PCFs which commonly have graphics).
See reply #1 above.
@DigitalXeron said:
Unfortunately, I don't have a Windows system with me to test your tool so I am unable to use it personally. (I use linux)You don't want to use it. Don't even try to get it running under Wine. Your sanity is at stake...
-
Don't try this without Spaghetti Code
@MasterPlanSoftware said:
@DigitalXeron said:
SpectateSwamp:
Some questions for you and your offering:
1. Can it search an entire hard disk of files? (files of which cannot be concatenated into one large file - such as legal documentation, files of different formats)
2. (Borrowing from someone else): Can it search using regular expressions? (Perl-compatible is better)
3. Since your tool doesn't have an index, can you give us some benchmarks on search speeds, CPU utilization and execution time on different types of searches?
4. Can it search through PDFs, PostScript, RTF documents and other formatted files? (I ask this as I wish to know if your tool can search though manuals in formats such as PDFs and other formatted documents that cannot be transformed into plaintext files such as PCFs which commonly have graphics).
Unfortunately, I don't have a Windows system with me to test your tool so I am unable to use it personally. (I use linux)
His replies will (sadly) be:
1. Blah Blah Blah you should combine all your files on your computer into one giant text file.
2. What are regular expressions?
3. It is very fast. But the code sucks. But who needs optimization? Vax. Something about the 80's.
4. You can export all of these files into text, and then they can be searched.
If he answers, that is what it will amount to. None of it will make any sense at all, and you will be left with a headache.
I was going to agree with MasterPlanSoftWare but that would be just wrong.
1. Right now the app is designed to merge/append every .txt or other text type files it finds on a drive or sub-folder. This doubling of size for some may run them out of disk space. A change or two to the code could make it do what you want. (Don't try this unless you have Spaghetti code). Right in the merge code when the input files are read and output to the merge file. Place the ever powerful "GOTO" to jump to the point right after the main logic read from the usual target input file. A check for this new search type here and there and the code should be able to do a search while in a psudo merge mode.
2. NoShow was added to the search after my sister got tired of seeing her X when we were testing the family album picture options. All my searches are "and" searches. The ss option is an "or" search.
3. Speed is limited to how fast the sequential text file can be moved from Disk. The faster the disk the faster the search. After the first pass on the file. Unless it is huge huge huge. The file will reside in memory. To show the last page of my notes file. The program reads from beginning to end counting the lines in the file. Then reads it a 2nd time (from memory) up to the number of lines in the file Less 30 or so and starts the display there. Showing the last page. Database files don't move data like a sequential dump. If you need even more speed, Then the program can run multiple copies of itself. Each searching a given section of the database. Extracting it so it can be merged and displayed by the Mother Search. Try this app out on a friends computer. 20,000,000 characters per second is about what this program can do. Faster disks mean faster search
4. With a minor builder app. If someone would write this simple vb program. Right now I can search and display pictures with this search. The first line being the detail line to search and the 2nd being the path to the picture. The routine would create 2 lines in the new text file. For each line of text in the originating odd ball format file. The 2nd line would be the path to the pdf file along with more info like page number. Instead of opening a picture. The program would be opening the pdf file with the originating software. Almost instantly. Very few changes would have to be made so that a new format could be added. If you have a friend with VB have him do it.
-
@Cap'n Steve said:
he also wrote the best bot I've ever seen (grammar not withstanding)
I take it you've never seen a properly trained megahal, then. Once you've been on an IRC channel with one for a while, you get the hang of spotting them. It's really quite an old piece of software - turns out that semi-comprehensible English isn't that hard to generate, and it's kinda convincing because most people who speak English just aren't very comprehensible to begin with. Algorithms like that one have generated scientific papers that were convincing enough to be accepted by conference reviewers.
(There's obviously a person supporting it and keeping it roughly on-topic)
-
Ok. First of all: The "comma" was invented for a good reason. It makes things so much more read- and understandable (esp. for people whose first language is not English - like me). But maybe grammar is like formatted text - no one really needs it, right?
1) Oh boy - if you know your algorithm is that shitty, why don't you fix it yourself? Please explain to me in which way GOTO is "ever powerful"?
2) That must be your way of asking "What are regular expressions", right? You don't have the slightest clue what DigitalXeron is talking about here, do you? REs have nothing to do with AND or OR searches...
3) Are you kidding? Oh, by the way: Isn't counting the rows some sort of indexing, as you later use the line number as a key for the position in the file you're displaying? Uh oh - gotcha!
4) This is indexing! Only that you're creating the index manually. What do you do for picture search? Create a text file with two lines per picture, one of them containing keywords and the other one containing the path? If that's not some sort of index I don't know...
So basically the way your "program" works is the result of the fact that a) you don't know how to index properly and b) you don't know how to automatically extract information from anything but a text file? That's so poor considering the fact that you've been working on it for - what - 10 years? So, if I'd like to index a PDF file I'd need to put a line in a text file providing the PDF's text as keywords. Priceless!
You should give other (real) Desktop Search programs a go. Just pretend to check out how crappy they are compared to your so called Desktop Search if that makes you feel easier. You may. However find. That. In fact they. Do a much better. Job than. Your thing. And they can be used. Intuitively.PS: Somewhere in your previous posts you stated that you've been successfully introducing people with no computer knowledge whatsoever to your "Desktop Search". You say they were impressed how "easy" things are with your program. Has it ever occurred to you that they - having no computer experience - might just not know better?
-
Great Software is easily customizable
@tdittmar said:
Ok. First of all: The "comma" was invented for a good reason. It makes things so much more read- and understandable (esp. for people whose first language is not English - like me). But maybe grammar is like formatted text - no one really needs it, right?
1) Oh boy - if you know your algorithm is that shitty, why don't you fix it yourself? Please explain to me in which way GOTO is "ever powerful"?
2) That must be your way of asking "What are regular expressions", right? You don't have the slightest clue what DigitalXeron is talking about here, do you? REs have nothing to do with AND or OR searches...
3) Are you kidding? Oh, by the way: Isn't counting the rows some sort of indexing, as you later use the line number as a key for the position in the file you're displaying? Uh oh - gotcha!
4) This is indexing! Only that you're creating the index manually. What do you do for picture search? Create a text file with two lines per picture, one of them containing keywords and the other one containing the path? If that's not some sort of index I don't know...
So basically the way your "program" works is the result of the fact that a) you don't know how to index properly and b) you don't know how to automatically extract information from anything but a text file? That's so poor considering the fact that you've been working on it for - what - 10 years? So, if I'd like to index a PDF file I'd need to put a line in a text file providing the PDF's text as keywords. Priceless!
You should give other (real) Desktop Search programs a go. Just pretend to check out how crappy they are compared to your so called Desktop Search if that makes you feel easier. You may. However find. That. In fact they. Do a much better. Job than. Your thing. And they can be used. Intuitively.PS: Somewhere in your previous posts you stated that you've been successfully introducing people with no computer knowledge whatsoever to your "Desktop Search". You say they were impressed how "easy" things are with your program. Has it ever occurred to you that they - having no computer experience - might just not know better?
Are you an English teacher or what. If you corrected people like that face to face. You'd be slapped silly. Oh you are already there.
GoTo and shitty, Try and make major changes to programs that arn't spaghetti. If it is well designed a well placed GoTo can do Magic.
Regular expressions. Didn't need em in 1970, never used in in 38 years. Still don't need them. What is "Gee Haw" mean that's an expression that was much more common and widly known. That's what I call regular expressions. Gee Haw to you. This is too much fun.
Counting lines. I know it's not efficient. But it is as clear as I can get when leaving the code for others. If it were way complex. Then changes couldn't be made easily. Good code can be changed easily by others. I think you'll find others will take and change this code.
4) indexing. In a way maybe so. But in a way that is completely understandable to others / users. That's why I can explain it to the common man.
Years and years ago 1974 My late friend Grant C. said If it was up to him; when designing systems he would do them all with text files. Corrections could easily be made to the data with editors. The data didn't need any unpacking of numbers etc to make them understandable. I'm just following through on this vision from my friend.
There are just too many different formats for me to even attempt to start handling them all. But now with other smart guys and gals like yourselves I don't have to worry about that do I. Search_pdf Search_oddball There should be a version to do everyone of these formats. Soon. I'm too busy doing video to worry about archiving formats I don't use. You do it.
New Point.
I have left some code where users can be Identified. Left in for a purpose but bypassed. Say you want to modify the code and restrict certain users to display only or to certain files. It's sooo easy. Great Software is easily customizable to fit any situation.
-
Gee Haw Explained.
The day this picture was taken Gee and Haw were used lots.
http://www.telusplanet.net/public/stonedan/scn03873.jpg
It's a dog team command. (left - Right) When you are teasing somebody. You are Gee Hawing them. Try saying that to this team and they would probably bite you.
Picture is 50 years old. Thats me and my cousin elaine. The driver is Wes Reid. Easy to find stuff like this when you have it cataloged.
-
@SpectateSwamp said:
Easy to find stuff like this when you have it cataloged.
Indeed. That's why I use the search function within Windows Vista which makes it possible to find pictures based on the tags embedded in pictures. And when I throw away Windows Vista, any other search will be able to index my pictures based on their tags.
Your "desktop search" can only search a text file where you had to manually say that scn03873.jpg contains a picture of you, your cousin and the driver.
-
Txt catalogue valuable reference.
@wooter said:
@SpectateSwamp said:
Easy to find stuff like this when you have it cataloged.
Indeed. That's why I use the search function within Windows Vista which makes it possible to find pictures based on the tags embedded in pictures. And when I throw away Windows Vista, any other search will be able to index my pictures based on their tags.
Your "desktop search" can only search a text file where you had to manually say that scn03873.jpg contains a picture of you, your cousin and the driver.
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine. The text is valuable and will always be. When I have my pictures numbered sequentially. I can do a visual check to see if any are missing. Data goes bad you know. How in the H would you tell if any file happens to go bad?. With this search the picture can be displayed for a few seconds in stretch mode. Then brought up in original size. In original it is possible to display large font text as to who is in the picture. Right on top. Oldie pictures with names. I havn't tried using the scrolling font feature there. But should. It's all there in the code. If it does something you don't like get a friend with vb and the 2 of your fix it.
-
@asuffield said:
@Cap'n Steve said:
he also wrote the best bot I've ever seen (grammar not withstanding)
I take it you've never seen a properly trained megahal, then. Once you've been on an IRC channel with one for a while, you get the hang of spotting them. It's really quite an old piece of software - turns out that semi-comprehensible English isn't that hard to generate, and it's kinda convincing because most people who speak English just aren't very comprehensible to begin with. Algorithms like that one have generated scientific papers that were convincing enough to be accepted by conference reviewers.
(There's obviously a person supporting it and keeping it roughly on-topic)
I call BS. This is clearly someone who has completely lost it.
-
@SpectateSwamp said:
@wooter said:
@SpectateSwamp said:
Easy to find stuff like this when you have it cataloged.
Indeed. That's why I use the search function within Windows Vista which makes it possible to find pictures based on the tags embedded in pictures. And when I throw away Windows Vista, any other search will be able to index my pictures based on their tags.
Your "desktop search" can only search a text file where you had to manually say that scn03873.jpg contains a picture of you, your cousin and the driver.
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine. The text is valuable and will always be. When I have my pictures numbered sequentially. I can do a visual check to see if any are missing. Data goes bad you know. How in the H would you tell if any file happens to go bad?. With this search the picture can be displayed for a few seconds in stretch mode. Then brought up in original size. In original it is possible to display large font text as to who is in the picture. Right on top. Oldie pictures with names. I havn't tried using the scrolling font feature there. But should. It's all there in the code. If it does something you don't like get a friend with vb and the 2 of your fix it.
Pictures go bad?
LMFAO
-
@SpectateSwamp said:
@wooter said:
@SpectateSwamp said:
Easy to find stuff like this when you have it cataloged.
Indeed. That's why I use the search function within Windows Vista which makes it possible to find pictures based on the tags embedded in pictures. And when I throw away Windows Vista, any other search will be able to index my pictures based on their tags.
Your "desktop search" can only search a text file where you had to manually say that scn03873.jpg contains a picture of you, your cousin and the driver.
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine. The text is valuable and will always be. When I have my pictures numbered sequentially. I can do a visual check to see if any are missing. Data goes bad you know. How in the H would you tell if any file happens to go bad?. With this search the picture can be displayed for a few seconds in stretch mode. Then brought up in original size. In original it is possible to display large font text as to who is in the picture. Right on top. Oldie pictures with names. I havn't tried using the scrolling font feature there. But should. It's all there in the code. If it does something you don't like get a friend with vb and the 2 of your fix it.
You have been challenged before to post your videos (note: with REAL screen capture software, NOT video of the screen) with the software you claim you are better than, side by side. You attempted to take on Notepad in the text searching capability. You failed. If your intent is to take on pictures and videos, by all means show us how. Show two identical searches of your HD with your search and a competitor's.
Otherwise, we all know you are full of shit anyway. Show us the proof, or go away.
-
@MasterPlanSoftware said:
@asuffield said:
@Cap'n Steve said:
he also wrote the best bot I've ever seen (grammar not withstanding)
I take it you've never seen a properly trained megahal, then. Once you've been on an IRC channel with one for a while, you get the hang of spotting them. It's really quite an old piece of software - turns out that semi-comprehensible English isn't that hard to generate, and it's kinda convincing because most people who speak English just aren't very comprehensible to begin with. Algorithms like that one have generated scientific papers that were convincing enough to be accepted by conference reviewers.
(There's obviously a person supporting it and keeping it roughly on-topic)
I call BS. This is clearly someone who has completely lost it.
I have a new theory :
Apollo mission trained monkey. Sent to Canada on a well deserved retirement.
-
-
@SpectateSwamp said:
Regular expressions. Didn't need em in 1970, never used in in 38 years. Still don't need them. What is "Gee Haw" mean that's an expression that was much more common and widly known. That's what I call regular expressions. Gee Haw to you. This is too much fun.
Hahahaha oh wow.
-
@SpectateSwamp said:
@wooter said:
@SpectateSwamp said:
Easy to find stuff like this when you have it cataloged.
Indeed. That's why I use the search function within Windows Vista which makes it possible to find pictures based on the tags embedded in pictures. And when I throw away Windows Vista, any other search will be able to index my pictures based on their tags.
Your "desktop search" can only search a text file where you had to manually say that scn03873.jpg contains a picture of you, your cousin and the driver.
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine. The text is valuable and will always be. When I have my pictures numbered sequentially. I can do a visual check to see if any are missing. Data goes bad you know. How in the H would you tell if any file happens to go bad?. With this search the picture can be displayed for a few seconds in stretch mode. Then brought up in original size. In original it is possible to display large font text as to who is in the picture. Right on top. Oldie pictures with names. I havn't tried using the scrolling font feature there. But should. It's all there in the code. If it does something you don't like get a friend with vb and the 2 of your fix it.
Now seriously, what is so different in your "app" then using a Word document :
For text :
1. Type name, keywords, etc.
2. Go to Insert->File and find a file you want to include in the "search"
For images/videos/music :
1. Type name, keywords, etc.
2. Mark them and press Ctrl-K and find an image/video you want
3. They turn to hyperlink, when you want to access them, you simply hold Ctrl and click
-
@SpectateSwamp said:
Then the program can run multiple copies of itself. Each searching a given section of the database. Extracting it so it can be merged and displayed by the Mother Search.
Haven't seen this part in your code, can you please post the latest version of the code ?
If we choose to help you, how can we work on the code, do you have some versioning system ? (CVS and SVN are probably a bad choice, you could try to write your own based on the DesktopSearch)
Is this project already registered on SourceForge (since it is open source) ?
-
@Nelle said:
@SpectateSwamp said:
Then the program can run multiple copies of itself. Each searching a given section of the database. Extracting it so it can be merged and displayed by the Mother Search.
Haven't seen this part in your code, can you please post the latest version of the code ?
If we choose to help you, how can we work on the code, do you have some versioning system ? (CVS and SVN are probably a bad choice, you could try to write your own based on the DesktopSearch)
Is this project already registered on SourceForge (since it is open source) ?
check out "backgrd" in the code. It's been a while since I used it. ie in a "match line" line #1 if there is a "backgrd==" and the path to a new exe then that job is started up in its folder ie c:\search\backgrd1\backgrd1.exe It has all it needs in it's own control file to run. The parent job can then start up another backgrd2 job (ie the next line pair) or wait for the first one to complete before it continues.
the code is at:
http://www.telusplanet.net/public/stonedan/source.txt
-
[Quote user="This Topic"]
[/Quote]
-
@SpectateSwamp said:
@Nelle said:
@SpectateSwamp said:
Then the program can run multiple copies of itself. Each searching a given section of the database. Extracting it so it can be merged and displayed by the Mother Search.
Haven't seen this part in your code, can you please post the latest version of the code ?
If we choose to help you, how can we work on the code, do you have some versioning system ? (CVS and SVN are probably a bad choice, you could try to write your own based on the DesktopSearch)
Is this project already registered on SourceForge (since it is open source) ?
check out "backgrd" in the code. It's been a while since I used it. ie in a "match line" line #1 if there is a "backgrd==" and the path to a new exe then that job is started up in its folder ie c:\search\backgrd1\backgrd1.exe It has all it needs in it's own control file to run. The parent job can then start up another backgrd2 job (ie the next line pair) or wait for the first one to complete before it continues.
the code is at:
So I found the part where the background process is called, but I can't seem to find the part where the background process communicates with the original one.
-
Here is the address of the shack. Here are directions to the shack. (So you know where to avoid. Or where to target the orbital lasers, if you're feeling ambitious.)
-
Spaghetti code with sauce and meatballs.
@Nelle said:
So I found the part where the background process is called, but I can't seem to find the part where the background process communicates with the original one.
You are right they don't communicate back with the original process. It is handy if you had 2 camera views of the same event and wanted to play both in sync. Just fiddle with the video start point on the 2nd video and move em around on the screen.
The navigation stuff is interesting too. Say the current line pair shows a splash screen with some selection instructions on it. The "control_files" array that are dated "23 November" code shows how. One can make a selection from 1 to 10 of various control files and their changing inputfile as you navigate along.
Way more than I want to do. But It could lead to something. Not that hard to implement either.
-
Multiple sessions are fun.
Oh Yeah. The background jobs are set so that they will run so long. The Parent executable is set to wait as long as it takes for the background job to display a random screen of scrolling text with jokes or motivational sayings then close itself out. Then on to the next line pair set that runs a background job to do something different. I used this for random political questions running continuously in a storefront window.
-
@SpectateSwamp said:
When I have my pictures numbered sequentially. I can do a visual check to see if any are missing.
If they're numbered sequentially, why don't you program SSDS to check if any are "missing"? Or even back-reference from your .txt file and see if the files being indexed are still there? Heck, even your "I can open files at random" thing could be used. Sit there long enough (and you do seem to have a LOT of free time on your hands) and eventually all the images would be popped up.
@SpectateSwamp said:
Data goes bad you know. How in the H would you tell if any file happens to go bad?.
We'll just use our technogeek MD5 and SHA hashes to validate the contents of our JPGs, and can find out if any are "missing" by doing a "dir /s *.jpg" and seeing if the counts match up.
@SpectateSwamp said:
Experience tells me that
I can teach someone how to use this program in an afternoon. They will
understand what is going on.Experience tells me I can sit someone down in front of a Vista/Google Desktop/Apple Spotlight search prompt and they will be able to find anything at all anywhere on their computer in less than 15 seconds. No tedious manual indexing of non-text files, no required concatenation of all text files into a single one. It "just works".
Yes, all those apps produce an index. And you know what? No one needs to know how the index works, or how it's built, or how it's stored. What matters is that the search works. With your "application" you force people to sit down and manually catalogue all their data. The other REAL search programs do that automatically. You keep harping about how much free time you have. I very much doubt you have any free time at all, as your ultimate time saver requires you to essentially waste your entire life building an index of your data.
In the mean time, everyone else with their non-SSDS programs will be out taking photos of birds, dog-sledding, etc.. and having FUN, because they don't have to sit in front of the computer as your crap forces them to.
You advocate that your program puts control in the hands of the user. I'll agree with that much. But only because you program forces them do to everything for themselves anyways. Given a choice between a lifetime of servitude trying to feed data into the bottomless maw of SSDS, or a couple minutes installing Goggle Desktop and letting it do all the work while I go outside and play, I'll choose Google Desktop (or any other search app that really does work).
You've previous claimed that SSDS removes the need for an operating system. Therefore I challenge YOU to remove Windows from your computer and have only your precious random videos and huge-ass text file, and the SSDS .exe file on the harddrive. Let's see how useful SSDS is at that point.
Time to put your operating system to rest and let SSDS free. Go ahead and make SSDS your operating system. You don't need Windows, all you need is SSDS. We'll be waiting here for you to report your results. Take your time, we're in no rush.
-
@SpectateSwamp said:
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine.
That's odd. I can take my pics - and just the pics - to any other operating system, and search them. I don't even need a textfile! The data's in the pic!
@SpectateSwamp said:
Data goes bad you know. How in the H would you tell if any file happens to go bad?
Suppose we accept data goes bad (unless your drive crashes or is being hugged by a magnet, it doesn't): that's why we have backups. There is no use in seeing if something went bad, if we don't have a backup
@SpectateSwamp said:
With this search the picture can be displayed for a few seconds in stretch mode. Then brought up in original size. In original it is possible to display large font text as to who is in the picture. Right on top. Oldie pictures with names. I havn't tried using the scrolling font feature there. But should.
That's odd that you need a textfile for that. Any other operating system these days can do without it. Again: the data is ín the picture.
@SpectateSwamp said:
It's all there in the code. If it does something you don't like get a friend with vb and the 2 of your fix it.
Yeah, uhmm, there is a problem with that: your code is very hard to understand. It is badly structured and comments do not explain anything. I'm therefore not sure where to start, íf i would decide to actually use your software. However, many other software is working perfectly for me, so I don't see the need for a text searcher with random video playing capabilities.
Don't forget: Open Source does not mean other people are gonna fix your bad code. Give them good code, and that might tick people off in extending the software.
One last suggestion: as you may start to feel, things changed since the '80's. One amongst others, is programming style and capabilities. It is probably not the best up-to-date book, but this book might help you in getting up to date with your VB6 programming habits:
http://www.amazon.com/Visual-Basic-6-Dummies-Windows/dp/0764503707
-
Interesting, SpectateSwamp seems to have a habit of digging his threads.
http://www.digg.com/tech_news/Why_is_Everybody_so_clueless_on_the_importance_of_Search/who
-
@SpectateSwamp said:
I can take my pics and my text file to any other operating system that can deal with jpg and txt. I think that's about all of them. If they have a simple text search function. And they should. Then I can find my files just fine.
So what your saying is that you can search text just fine without DesktopSearch?
-
Desktop Search more important than your Browser
@tster said:
So what your saying is that you can search text just fine without DesktopSearch?
Text is similar to Open Source in that respect. Nobody has a gun to your head and that this app searches text far better than the rest.
Everybody except me has been able to survive without this search. It's an accumulation of things that make this search so useful.
Instant notepad, no sql, no indexing, search large files instead of a spell checker. Customizable. Portable, Random....
That makes this search, the only real app, the man on the street needs. Forget all the other applications and their confusion.
I'll teach them everything this Desktop search does in a half day and their introduction to computers in done. Complete.
Can't do that without this simple yet powerful tool
-
@wooter said:
That's odd. I can take my pics - and just the pics - to any other operating system, and search them. I don't even need a textfile! The data's in the pic!
Putting data in pics where it is another step away from the user adds to the confusion. Just another reason for the man on the street to fear computers. Having to know words like metadata. It takes a lifetime to learn computing your way. If you ever really know it.
@wooter said:
Suppose we accept data goes bad (unless your drive crashes or is being hugged by a magnet, it doesn't): that's why we have backups. There is no use in seeing if something went bad, if we don't have a backup
Bad might have been a bad choice of words. Missing is probably better. When you have lots of video and pics like I do. Long unwieldy file names make things almost unmanageable when looking at file names. I can spot one missing in a sequence easily. I'm not sure I want to look at 5000+ family album pictures file name lists any other way that scn00234.jpg scn00235.jpg. Just easier to keep organized. I do keep lots of copies of the family pictures and related text file.
@wooter said:
Yeah, uhmm, there is a problem with that: your code is very hard to understand. It is badly structured and comments do not explain anything. I'm therefore not sure where to start, íf i would decide to actually use your software. However, many other software is working perfectly for me, so I don't see the need for a text searcher with random video playing capabilities.
1 simple program written by a dumb sob. with flowcharts for the hard parts.
There are lots of print statements you could re-activate. Jam jam jam the code. You'll quickly see what it is doing.
@wooter said:
Don't forget: Open Source does not mean other people are gonna fix your bad code. Give them good code, and that might tick people off in extending the software.
Open Source - Maybe someone will port it to Mac or Linux. I might even try a Mac again if that happens.
The code works just fine and does everything I need. There might be something yet to be done, that will excite you. Have at it.
Navigation, Slow motion, Freeze frame. And a great way to really bug geeks.
No need to change it right away. Taking a peek at the code can be enough for now.
-
The Brilliant SwampyBean
package DesktopSearch;
public class SwampyBean {
private String Swampy = "ClueLessNess";
public String getSearch(String Swampy) {
return 1970sBASICImplementationOfgrep(Swampy);
}
}
@SpectateSwamp said:Instant notepad, no sql, no indexing, search large files instead of a spell checker. Customizable. Portable, Random....
Instant Notepad? Ooh - does it edit & save files as well?
No SQL? Great - I never did like ordering, selective searching or those complex 'views' everybody talks about.
No indexing. Top. I always search through and re-index my files each and every time too - that's why I use perl :)
Sirch Larj Fyles insted of a Spill chucker. This is genius - why use a dictionary file when you can just use every bit of text you've ever written? That way, you'll never have to type words that you've never typed before! And that auto-correct nonsense always annoyed me - that's how we ended up with Clippy.
Customizable. What - you mean anyone can just change the source code? Great! Now then - where's that book of "How To Write Proper Arcade Games in Basic on Your IBM 8086 PC"...
Portable Excellent - 'cos what I really want is to run it on my custom OS/2 Betamax video indexing system - I've got all my Duran Duran videos classified by height of Simon's hair.
Random. Indeed.
Sorry for feeding the troll, but it's Sunday afternoon and I'm bored...
-
@Benn said:
Sorry for feeding the troll, but it's Sunday afternoon and I'm bored...It's ok, just use the right tag....
-
Desktop Search Nuts WANTED
@Benn said:
package DesktopSearch;
public class SwampyBean {
private String Swampy = "ClueLessNess";
public String getSearch(String Swampy) {
return 1970sBASICImplementationOfgrep(Swampy);
}
}
Instant Notepad? Ooh - does it edit & save files as well?
No SQL? Great - I never did like ordering, selective searching or those complex 'views' everybody talks about.
No indexing. Top. I always search through and re-index my files each and every time too - that's why I use perl :)
Sirch Larj Fyles insted of a Spill chucker. This is genius - why use a dictionary file when you can just use every bit of text you've ever written? That way, you'll never have to type words that you've never typed before! And that auto-correct nonsense always annoyed me - that's how we ended up with Clippy.
Customizable. What - you mean anyone can just change the source code? Great! Now then - where's that book of "How To Write Proper Arcade Games in Basic on Your IBM 8086 PC"...
Portable Excellent - 'cos what I really want is to run it on my custom OS/2 Betamax video indexing system - I've got all my Duran Duran videos classified by height of Simon's hair.
Random. Indeed.
Sorry for feeding the troll, but it's Sunday afternoon and I'm bored...Troll bored too. Very bored. Need more swampies.
The thing is with these posts. I learn a lot. Every time.
The source works for the PC. Now I just need some Mac, Linux and Java desktop search enthusiasts to take up the challenge. I'll help all I can!
With Spectate Swamp Search on all these platforms. I could move all my data from PC to Mac to Linux to the Internet and back again in hours. Now that is portability. Know any Search Nuts? Don't need all that many
My other next to favorite forum Locked the similar post. Troll cornered. No place to go.
-
@SpectateSwamp said:
The thing is with these posts. I learn a lot. Every time.
Sorry mate, but the thing is, you don't. My first post was a reasoned attempt to show you this, but still you persist. There are lots of phrases used here and elsewhere - Re-inventing the Wheel, Not Invented Here etc. - that all apply in this situation, but what you don't seem to grasp is that for most people here (I should think), the quality of code is directly related to the quality of thinking about the particular problem. I don't need to know anything about your particular solution to the Desktop Search Problem; all I need to do is look at your code , which I have, as I presume, have most people contributing to this thread. And like them, I laughed / cried / had all the other reactions of a professional artisan looking at the worst apprentice piece he'd ever seen. I understand that this brave new object-oriented dynamic online XML world all looks very frightening to someone who learnt to tame a Vax, but please, I beg you, re-read this thread from the very beginning and attempt to understand what every single poster is urging.
-
Swampies WANTED
@Benn said:
@SpectateSwamp said:
The thing is with these posts. I learn a lot. Every time.
Sorry mate, but the thing is, you don't. My first post was a reasoned attempt to show you this, but still you persist. There are lots of phrases used here and elsewhere - Re-inventing the Wheel, Not Invented Here etc. - that all apply in this situation, but what you don't seem to grasp is that for most people here (I should think), the quality of code is directly related to the quality of thinking about the particular problem. I don't need to know anything about your particular solution to the Desktop Search Problem; all I need to do is look at your code , which I have, as I presume, have most people contributing to this thread. And like them, I laughed / cried / had all the other reactions of a professional artisan looking at the worst apprentice piece he'd ever seen. I understand that this brave new object-oriented dynamic online XML world all looks very frightening to someone who learnt to tame a Vax, but please, I beg you, re-read this thread from the very beginning and attempt to understand what *every single poster* is urging.True true. But bad code can run. Even with Goto's
What is needed is 1 or 2 people who are willing to expand search to platforms other than the PC. Then my data becomes truly portable.
If you don't have VB that's ok. Just email me the source and I'll build a new exe for you and email it back. It doesn't have to be anything more than a name change or prompt fix. I'll put the code in and return you the exe. If someone would fix that nasty startup problem for me I'd appreciate it. Should just close it out complete Maybe?? I think I could use 8 or 10 full time swampies more than that. I'd be swamped.
-
@SpectateSwamp said:
The source works for the PC. Now I just need some Mac, Linux and Java desktop search enthusiasts to take up the challenge. I'll help all I can!
For Mac and Linux, here you go:
#!/bin/sh
# (rev. 1) Get rid of -E: we don't want no stinkin' regex.
grep -d recurse $@ $HOME
Oh wait, I forgot one major feature.
#!/bin/sh
grep -d recurse $@ $HOME |tee /tmp/pick
# (rev. 1) Get rid of -E: we don't want no stinkin' regex.nlines =
wc -l /tmp/pickline =
expr \( 1 + $RANDOM \) \* $nlines / 32768# (rev. 2) Don't forget to open something randomly.
open "
head -n $line |tail -n1 |cut -d: -f1"
