Alexandre Guimaraes wrote: > I have a huge amount of paper with application notes, printouts from > list messages, data sheets, interesting projects, etc, etc... > > I think most of the professionals on the list professionals on the > list also have. How do you organize the mess ?? I tried so many ways over > the years that it is hard to describe all. I have the opportunity to hire > a > person to organize the papers or make it all digital by scanning what is > not > in electronic format and grabbing the PDF's when it is still available. Is > there any nice way to do this ? How to "decide" wich "categories" to have > ? > Any software to do searchs after it is all in ? > > Sorry for asking so many questions but I really hate to reinvent the > wheel when someone probably has already done something similar... These are good questions, that everybody struggles with. THE PROBLEM In my opinion, the goal here is to minimize the overall effort, which consists of two parts: 1. The effort required to enter an item into the system. 2. The effort required to find/retrieve an item. THE SOLUTION The best advice I can give you, is to buy the 'Getting Things Done' book by David Allen. It deals with precisely the problem you've described: organizing the mess. Prior to reading the book, my desk was an absolute mess: my coworkers had to put papers on my chair to make sure they didn't get lost in the morass! It took me a while to go through the mess, paper by paper, to find a particular document. Since I've started following the simple procedures given in the book, several things have happened: - My desk and drawers are neat and organized. - I can find the document I need, in under a minute. - I can think a lot more clearly than before. The big difference between the GTD method, and others (like "the 7 habits") is the bottom-up approach: you put the low-level details in order first, and then you figure out your higher priorities. The most useful things I got out of the book are: - A simple, alphabet-based filing system. _No_ categories, one folder per hanging file, with files arranged in alphabetical order. - Labeler is *essential* to making the above system work. Hand-written labels, or labels printed on an inkjet/laser printer just don't cut it. The cheap ($20) Brother PT-80 labeler has the best cost/performance ratio (especially when you consider the cost of tapes). - The inbox that gets processed and emptied every day, with things either done, filed away, or delegated. - The tickler file, which I am *just* starting to use. The key to the GTD success is minimizing the effort you put into the system, while getting as much as possible out of it. HARDWARE PROJECTS For hardware projects, we maintain documents in two forms: electronic and paper. = Electronic = Datasheets, schematics, project notes, utilities specific to a project -- basically, anything related to the project that is in electronic format, go into a folder on the local fileserver (//server/proj_hw/myproject). The project folder has subfolders for datasheets, schematics, manufacturers files, etc. Those can also have subfolders if needed: for example, datasheets can be broken down into ics, passives, transistors, etc. I can normally find what I'm looking for by simply browsing to it. In rare cases, I use the built-in Windows search function, to find a document by its filename. = Paper = Everything in paper form, including frequently used pages from datasheets (footprints, tables, etc) and handwritten notes, go into a paper folder. When not in active use, the folder is put in a filing cabinet (and can be retrieved in a matter of seconds). THE BOTTOM LINE When considering an organizing system, make sure that it takes the least amount of effort possible to create, maintain, and use the system. CRITIQUE Gerhard suggested that hierarchical categories don't make a lot of sense, and that one should rely on a search engine. This surely minimizes the effort to create and maintain the system, but it can make finding things very hard. If you don't have any categories at all, you may find yourself running "global" searches, which tend to take a long time, and/or produce too many results if the search terms aren't specific enough. Bob Axtell's project binder is similar to what we use for our projects, but it sounds like his takes more effort to create and maintain. We produce only the bare minimum of printed documentation (e.g., one page from the datasheet, rather than the whole thing), and put all of it in one folder. Related pages are stapled together. Eoin's electronic system sounds almost identical to ours. I would emphasize the "YYYY MM DD" format for files/folders that are not on CVS, but that shouldn't be overwritten. My examples include CAD files, successive versions of files (ex: contracts, company logo) that get e-maield back and forth, folders for storing images from the digital camera. Wiki is a good idea for "evolving" documents, but I don't see it working as an organizing system for external paperwork. Cedric described a system that sounds like a great idea, but I don't like the extra step that the database creates, and the fact that everything is filed using a number (which are more difficult to remember than words). To file something using Cedric's system, you have to: 1. Open the database. 2. Fill out the fields (created, revised, author, description, number). 3. Put the number in the document. 4. File the item. To retrieve this item, you have to: 1. Open the database. 2. Enter the keyword(s). 3. Get the number (better write it down!) 4. Use the number to find the paper file. As Cedric has pointed out, the database is the key to this system, but I think it is also its bottleneck and Achilles' heel. Without access to the database, everything unravels. I find it easier to simply label the paper folder ("Lunar Strobe"), and put it in the filing cabinet (after "Lucent Technologies"). Retrieval is just as easy: open the drawer, and find the "Lunar Strobe" under "L". As an added benefit, the labels transform the filing drawer into a neat, self-updating table of contents. Of course, it makes sense to separate different things by filing cabinet. If you have lots of product samples, put them in their own filing cabinet. Same goes for product manuals. This system has actually been field tested, works great. Besides being more efficient than a computer DB-centric system, it is also more robust. ---> HOWEVER, we are currently considering a database system similar to Cedric's, for our component drawers. There are two reasons: 1) it is not easy to add new drawers in alphabetical order, and 2) we need to be able to quickly determine the quantity for a large number of components (as in a BOM). The numbers in this system would correspond to rows and columns (something like RRR:CCC). In this case, the cost of DB access is well justified by the time savings in other areas. Lastly, I would advise you against scanning all your existing papers into PDFs, unless you're really short on physical space, or have some specific need that only PDF can fill (such as being able to e-mail the document, or use the text search function). Good luck, Vitaliy -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist