Linux.com

send just a file : curriculum_vitae

#

But, it's pretty likely that there could be two Bill Johnsons competing for the same position. So, maybe it should be bill_h_johnson_resume. But, H is a pretty common middle initial. Probably I should spell out my full middle name, as in bill_howard_johnson_resume. Even so, it's not out of the question that there could be two Bill Howard Johnsons. Perhaps a short description would be in order: bill_howard_johnson,BSCS_from_stanford_university<nobr>_<wbr></nobr> with_a_minor_in_mathematics_spent_a_few_years_at_<nobr>p<wbr></nobr> ropietary_software_dot_com_and_am_now_working_on_<nobr>o<wbr></nobr> pen_source_projects looks better. However, I might want to send an updated copy to somebody. And s/he might not want to overwrite the previous copy. I'd better add a date and revision number, then, just in case: bill_howard_johnson,BSCS_from_stanford_university<nobr>_<wbr></nobr> with_a_minor_in_mathematics_spent_a_few_years_at_<nobr>p<wbr></nobr> ropietary_software_dot_com_and_am_now_working_on_<nobr>o<wbr></nobr> pen_source_projects_20040721_191733gmt_r000000

Perfect! I'll use that naming convention for all my files from now on!



Seriously, it's probably better to let the recipient give the file a name that makes sense to him/her, rather than guessing or making an assumption about what s/he wants.

#

I like your file name but I wonder if it doesn't violate some name length restriction.<nobr> <wbr></nobr>:-)

IMHO, the person who's putting themself in the position of receiving all these sorts of files should bear the burden of keeping the file names straight. Would it really be that much of a chore to save the file as:

bill_h_johnson_resume.20040721-1414.sxw

or even

resume.20040721-1414.sxw

or whatever the date and time was when you received or read the file?

I think roblimo's just having a tough day.

#

This problem is very serious. One of the things I frequently do when sending a frequently revised file to someone is to copy it to an archival name. In other words:

<tt>$ cp resume.ltx Resume-Anonymous-Reader-2004-07-21.ltx</tt>

Then I send it. That solves two problems. First, it makes the contents of the file pretty obvious (that's a LaTeX copy of my resume folks). Second, I don't edit or overwrite the archival copy. If I get the interview, I know which version of my resume they got. I have several that emphasize different skill sets.

#

So you didn't like the article, IOW. Or you just missed the whole point of it (common sense is a new concept for you?)

#

Two suggestions:

When the date is important, use the ISO date format of yyyymmdd. It sorts nicely, and avoids the month/day difference between US and European conventions.

When using numbers, keep the number of digits constant. In other words, use leading zeros. Files named part01, part02 sort and read better than part1, part2 when you get to part10.

#

been preaching this for years, the users of mine
who do it, like it. They PREPEND their files
20040721a_filename.ext

and they have NO trouble finding what they
are looking for without regard to where it
is, or what the filesystem THINKS the
file date is.

the "a" goes from a to 9, giving plenty of
room for revisions.

#

Amen!!! The ISO date formate RULES!!! I always thought that I was the *only* person to use this format religiously, and I'm glad to see that someone else does it too.

#

avoids the month/day difference between US and European conventions.

One minor correction. No-one outside of the US uses the US-style. The difference is not between US and Europe it's between the US and THE REST OF THE WORLD!

#

Amen, brother. After growing up in Europe and coming to the States in '88 it took me years to be able to parse the dates right. When I date checks or other written stuff I still find myself writing in the propper dd/mm/yy format.

And don't even get me started on why the US isn't on the metric system.

(Hint: people here are to stupid and companies are to stupid and greedy).

#

...well, if you think like a geek, and can remember commands but not the current date:

$( date +%Y%m%d )-descriptive-filename

Yes, I do this too. Both at the command line and in scripts -- where I'll add hour, minute, and second. And a random hash to avoid collisions if I'm in belts-and-suspenders mode ($RANDOM).

Some might ask why not just use the file's inherent timestamp. Problem is that that's an attribute of the file, and when it was created or modified last, not the document. This used to show up frequently on website search engines which tracked documents by file date. Clearly there was a lot of wholesale maintenance going on, because years-old articles would frequently show up among "current" results.

And it would be beneath me to mention InfoWorld Electric by name.

-- Karsten M. Self

#

Especially the binary newsgroups.

They post one hundred pictures of Pam Anderson.

Each one is named differently. "pa.jpg", "patits.jpg", "Anderson.jpg". Try sort those into some sort of order. Especially when you have someone else uploading, say, "Anderson.jpg" which happens to be of Laurie Anderson.

Or each one is named "1.jpg", "2.jpg", etc. That's nice - not. In my images directories, I have subdirectories for each person's image, with both name and number in the file name. In my download directory, however, Pam's "1.jpg" gets right next to Jennifer Aniston's "1.jpg". Not terrible useful when I want to do a mass rename of all the Pam Anderson downloads separately from the Jennifer Aniston downloads using a bulk rename utility.

In other words, morons, try to use intelligent file names that both identify the file and allow related files to be sorted together.

#

Good thoughts. Simple and clean; I like that. Everyone should name files by names that mean things. It's a pain to run through file names that are varients of some personal note with numbers for revisions.

The comment about ISO date stamping the name is a good idea. The original article also suggested good ideas. Basically, think a little and it makes life easier for everyone.

If I get bill_h_smith_resume.sxw, it means something. When shows up in a bin somewhere, at least I know where to start and should their be another bill_h_smith_resume.sxw, it limits my workload. Tagging filenames with ISO standard dates can help you and others more easily determine which file was made when, without looking through attributes.

It simply makes life easier when others think a little about your needs. Let's face it. We have a hard enough time reading people's mind due to poor grammar and language in email, letters, notes or even phone conversations. It's a good idea to think about communication now and then just to be considerate. I commend you for suggesting people do just that.

#

"We have a hard enough time reading people's mind due to poor grammar and language in email, letters, notes or even phone conversations."

"...at least I know where to start and should their be another bill_h_smith_resume.sxw..."

You're right about the poor grammar and language... their's just too much of it.<nobr> <wbr></nobr>:)

#

Rob, your suggestion would be a nice feature for Mozilla, or an extension for Thunderbird: an ability to rename an attachment when sending, thus not requiring the user to rename it on save at the other end. I'm not sure how many users would employ it though?

Hopefully, your suggestion would be given more consideration than my suggestion that a user be permitted to delete the 5 meg attachment from the e-mail after he saves. Such would greatly reduce the size of archived e-mail.

#

KMail has that feature: After attaching a file you can right-click it in the attachment list and use the "Properties..." dialog to change file name and encoding.

In addition KMail can also remind you if it looks like you forgot to add an attachment again. Another great feature every email client should have.

#

Nah. I agree and disagree.

Putting the file index.html on a P2P network is plain stupid. It happens though and it is the standard filename. In such case it is most likely a misconfiguration or there is nothing wrong at all except it blurs up the search. A whole website shared could also be shared as tarball.

My point is that you blame the person sharing. Such blames aren't always valid. I do agree a person who mails you a CV should attach it CV_name or something like that however if he/she fails you have the option to save it to that filename instead. Not a big deal. Wether its you or him/her who's renaming it always costs the same time unless you spread the data futher without renaming assuming the person who send it to you is the original author.

Then come the<nobr> <wbr></nobr>.txt and no_extention arguments, and the arguments people want to share files anonymous on P2P without their ''name'' or something similar which shows its uniqueness.

In the end there's no dogmatic point while in your situation i can imagine your problem. Your point is one to keep in mind, i agree, but not a golden rule. It has a better place as ingredient for pragmatic thinking.

I do totally agree with the person who's mad at Usenet here above. Then again, hords of Usenet people use YEnc, which shows on their ignorance.

#