shimself
Member of DD Central
Posts: 2,563 
Likes: 1,171
|
Post by shimself on Jan 2, 2015 11:53:13 GMT
When downloading data from a site which version do you select? (Just to bias the results I prefer .xls, I find csv inconsistent and it can be hard work, and being cheapskate I have to go via a convertor to use .xlsx on my old version of ms office)
If we get a clear result then it will be a good steer to the various platforms
|
|
|
|
Post by tybalt on Jan 2, 2015 12:18:38 GMT
.XLS I am reduced to mostly .xlsx on my home PC but prefer .XLS
|
|
pikestaff
Member of DD Central
Posts: 2,189
Likes: 1,546
|
Post by pikestaff on Jan 2, 2015 16:34:05 GMT
For downloading data it makes no difference to me. I will take whatever I am given. Having said that, .csv is more compact and should be future-proof.
I have Excel 2007, so I use .xlsx files every day and I prefer them for actual work - mainly because the file format is more efficient than .xls, which can make a massive difference to the file size.
When I upload files to forums I usually do so in .xls format for backwards compatibility, but this is not always possible. This forum has a maximum file size of 1 Mb. I recently uploaded an .xlsx file with embedded charts, which only just scraped under the limit at 965kB. The .xls version of the file was over 9 Mb, so no chance!
|
|
dave
Member of DD Central
Posts: 86
Likes: 38
|
Post by dave on Jan 2, 2015 16:34:28 GMT
CSV for me, much more chance I will still be able to read the file in 20 years time.
I could not have written my De-Shrapnelator for my reports from AC without CSV data (and Perl).
Dave
|
|
|
|
Post by tybalt on Jan 2, 2015 16:41:43 GMT
Now this will date me I still have attacks of nostalgia for MicroModeller and MasterModeller. I had not realised .xlsx was better at data storage than .xls
|
|
|
|
Post by Ton ⓉⓞⓃ on Jan 2, 2015 16:55:07 GMT
CSV for me, much more chance I will still be able to read the file in 20 years time. I could not have written my De-Shrapnelator for my reports from AC without CSV data (and Perl). Dave De-Shrapnelator you should patent that.
|
|
mikeb
Posts: 1,072
Likes: 472
|
Post by mikeb on Jan 2, 2015 18:59:53 GMT
I could not have written my De-Shrapnelator for my reports from AC without CSV data (and Perl). Dave De-Shrapnelator you should patent that. Ahem. That's De-Shrapnelator [tm], I think you'll find. You should know better (seeing as I first used the phrase in a PM to you)  I voted for .CSV -- it's a standard. It's open. It's human readable and easily processed with simple text-based tools (grep, sed and their ilk) to do adhoc data processing. XLS is a Microsoft format, standard by sheer force of number of users, nothing more. XLSX is like a chopped up XLS with bits of side salad thrown into in a .ZIP file. It may be great for embedding charts, graphs, formatting, pretty links etc. etc. but for raw data I dislike it. The format is liable to change and breakage on a whim. Maybe ODS should also appear on the poll (Open Document, Spreadsheet) Before OpenOffice/LibreOffice, such things [XLS/XLSX/DOC/DOCX] were always a bit of a pain, and various converters didn't always do a great job of pulling out the data correctly.
|
|
pikestaff
Member of DD Central
Posts: 2,189
Likes: 1,546
|
Post by pikestaff on Jan 3, 2015 9:07:08 GMT
...Before OpenOffice/LibreOffice, such things [XLS/XLSX/DOC/DOCX] were always a bit of a pain, and various converters didn't always do a great job of pulling out the data correctly. But now that we have them there's no need to worry any more. My only excuse for still using MS Office is I've got it and I can't be bothered to learn something else, even if the learning curve is small. My children use LibreOffice and I may well switch next time I change my PC.
|
|
Steerpike
Member of DD Central
Posts: 1,978
Likes: 1,687
|
Post by Steerpike on Jan 3, 2015 10:21:13 GMT
De-Shrapnelator you should patent that. Ahem. That's De-Shrapnelator [tm], I think you'll find. You should know better (seeing as I first used the phrase in a PM to you) I voted for .CSV -- it's a standard. It's open. It's human readable and easily processed with simple text-based tools (grep, sed and their ilk) to do adhoc data processing. XLS is a Microsoft format, standard by sheer force of number of users, nothing more. XLSX is like a chopped up XLS with bits of side salad thrown into in a .ZIP file. It may be great for embedding charts, graphs, formatting, pretty links etc. etc. but for raw data I dislike it. The format is liable to change and breakage on a whim. Maybe ODS should also appear on the poll (Open Document, Spreadsheet) Before OpenOffice/LibreOffice, such things [XLS/XLSX/DOC/DOCX] were always a bit of a pain, and various converters didn't always do a great job of pulling out the data correctly. CSV is a standard? Not in my experience, use and abuse of quotes, embedded quotes, commas, and embedded commas is rife, and I have never found a certified or universally agreed definition. For example, one can import a CSV file in to EXCEL and then export the same data to CSV and the files are different. Also, as the format incorporates no reliable formatting information one is often at the mercy of the client import/export "wizard". For example a column containing quoted string data will be interpreted by EXCEL as numbers and may be presented in exponential format. Nevertheless, CSV is compact, can be read in Notepad, and can usually be made to work satisfactorily. I used OpenOffice and later LibreOffice for some years but eventually returned to MS Office because MS EXCEL is just better, and MS Office is much more widely used and understood. Exchanging workbooks and documents is not straightforward and reliable either in ODS/ODF or DOC/XLS/DOCX/XLSX. I still have workbooks that lost all the formulae when transferred from one package to another and I never found a simple and reliable document protection system that I could use to transfer ODS/ODF files to other parties not using OO that did not require them to install software.
|
|
shimself
Member of DD Central
Posts: 2,563
Likes: 1,171
|
Post by shimself on Jan 3, 2015 15:55:32 GMT
Now this will date me I still have attacks of nostalgia for MicroModeller and MasterModeller. I had not realised .xlsx was better at data storage than .xls .xlsX is zipped.
|
|
shimself
Member of DD Central
Posts: 2,563
Likes: 1,171
|
Post by shimself on Jan 3, 2015 16:03:50 GMT
CSV for me, much more chance I will still be able to read the file in 20 years time. ... Dave My excel (and open office) will open Lotus123 which is over 30 years old (!). I don't imagine .xls will ever become unreadable direct into whatever we call a spreadsheet in 20 years time,
|
|
|
|
Post by westcountryfunder on Jan 3, 2015 18:12:07 GMT
I've voted for CSV, mainly because I find that more-often-than-not spreadsheets in .xls format are rarely properly formatted in the colums/rows/cells. If I am going to have to mess around with the download to make it useful then I would prefer to start from CSV.
For example consider the download for "MLIA Statement Summary" from AC, specifically the .xls version. It would be helpful if column A was formatted as a number. Similarly column E should be a number or currency to two decimal places. I'm not sure about column B. Format wise, it doesn't seem to be a date or a time.
I'm using LibreOffice on Linux Mint, but I doubt if that has any bearing on the formatting issue.
|
|
|
|
Post by longjohn on Jan 3, 2015 18:39:19 GMT
When downloading data from a site which version do you select? (Just to bias the results I prefer .xls, I find csv inconsistent and it can be hard work, and being cheapskate I have to go via a convertor to use .xlsx on my old version of ms office) If we get a clear result then it will be a good steer to the various platforms Have you installed the Office Compatibility Pack? This installs an add-on into Office 2003 that will natively read Office 2007, 2010 and 2013 files in Word, Excel and Powerpoint into Office 2000 and 2003 and save them properly too. It's free as well. MS Office Compatibility PackJohn
|
|
|
|
Post by longjohn on Jan 3, 2015 18:48:31 GMT
I've voted for CSV, mainly because I find that more-often-than-not spreadsheets in .xls format are rarely properly formatted in the colums/rows/cells. If I am going to have to mess around with the download to make it useful then I would prefer to start from CSV. For example consider the download for "MLIA Statement Summary" from AC, specifically the .xls version. It would be helpful if column A was formatted as a number. Similarly column E should be a number or currency to two decimal places. I'm not sure about column B. Format wise, it doesn't seem to be a date or a time. I'm using LibreOffice on Linux Mint, but I doubt if that has any bearing on the formatting issue. Are you sure that the file you are downloading is in a genuine .xls format? I've worked on various web sites where files have been exported as .xls when in fact they were bog standard .csv but saved with the .xls extension. They open in Excel without any notification of the erroneous file type and no column formatting. John
|
|
|
|
Post by westcountryfunder on Jan 3, 2015 19:12:05 GMT
I've voted for CSV, mainly because I find that more-often-than-not spreadsheets in .xls format are rarely properly formatted in the colums/rows/cells. If I am going to have to mess around with the download to make it useful then I would prefer to start from CSV. For example consider the download for "MLIA Statement Summary" from AC, specifically the .xls version. It would be helpful if column A was formatted as a number. Similarly column E should be a number or currency to two decimal places. I'm not sure about column B. Format wise, it doesn't seem to be a date or a time. I'm using LibreOffice on Linux Mint, but I doubt if that has any bearing on the formatting issue. Are you sure that the file you are downloading is in a genuine .xls format? I've worked on various web sites where files have been exported as .xls when in fact they were bog standard .csv but saved with the .xls extension. They open in Excel without any notification of the erroneous file type and no column formatting. John Interesting! I have noticed what I have described not only on AC but other sites such as FC. With AC there is the option to download either .csv or .xls. The resulting spreadsheets contain the same information but they don't look the same. Given what I have described, the .csv version is easier to work with.
|
|
