Any compelling reasons to use "odt" files instead of "doc" files?

26 replies [Last post]
Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
I used Word for many years before Microsoft ruined the product with its ’07 release. As such, I have hundreds of “doc” files. Now I use LO Writer for most things and SoftOffice TextMaker for documents I write in German. I have both word processors set to always save as MS Word ’97/00/XP/03 “doc” files. It’s the standard for all my files.

My previous standard was WordStar. Fortunately, I found a program that would automatically convert all my old WordStar files to “doc” ones, preserving the time and date of all of them. I was concerned a time may come when it was impossible to access those files.

However, there was a time in-between WordStar and Word when I used Lotus WordPro. I’m lamenting having used the default file “lwp” format. I haven’t found anything that will auto-convert them to doc files like I did with WordStar ones. For now WordPro still runs on my PCs. If I got really ambitious, I could open each one in WordPro, save it as a “doc” and then use software for changing its time and date back to the time when it was written. Yikes. What a bunch of work. Anyway, that’s the reason for the “doc” standard.

My documents are all simple. I only use Courier New or Times New Roman and only one or the other in a single document. I don’t use any tables or pictures. With this in mind, is there any reason I should use odt files? It looks like they’ll be around, unlike lwp files.

So far the doc files I’ve created work seamlessly from LO Writer to SoftOffice TextMaker. On the other hand, TextMaker does have the ability to make and read odt files. If I wanted, I could make that filetype my standard.

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
I can only recommend ODT
Tom, I would never recommend DOC over ODT as the former is a binary format. When a binary format file gets corrupted, you throw it out and start again 99 times out of 100. At least with a ZIP file containing XML (i.e., ODT) you are more likely able to recover pieces. The other problem with using any non-native format is that you are constantly at risk of the feature being deprecated or impaired in some way. Even in LO features can get turned off with each release e.g., v4.0, v4.1, etc.
sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
I agree with Oweng. A huge
I agree with Oweng. A huge advantage of the ODT format is that ODT is basically like a ZIP archive that contains, among other things, the text portion of the document in XML format. So if for some reason all version of LibreOffice and OpenOffice.org all mysteriously disappeared from the universe, at least you could unzip them with almost any archiving program and read the XML file. In your case, since you don’t have special formatting, the XML formatted text would be quite readable. In fact all the rest of the document is also controlled by XML. It’s always nice to be in control of your data and not depend on a single 3rd party company.

You can use LibreOffice to batch convert all of your DOC files into ODT, or even into PDF if you want to. Here are several ways:

___________________________________
LibreOffice 4.4, Ubuntu 15.04

Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
Thanks to you both. Those
Thanks to you both. Those are some compelling reasons to switch to the odt format. I wouldn’t even necessarily have to convert all my old doc files over. Doc is obviously going to be around for a while, and all my writings are extremely well backed up. I could just use odt from here on out.

The big question is what to do about my Lotus WordPro and my WordStar for Windows files. Lotus and WSWin both had their own formats. I was disappointed the conversion software that I used to convert all those WordStar for DOS files neither supported WordPro nor WSWin conversion. WordStar for Windows was not the same program as WordStar for DOS. WordStar International bought a program written by another company and then programmed in many WordStar features. WSWin used a completely different format than WS-DOS. And this format, known as “wsd,” was terrible and inefficient. It was way more wasteful and bloated than doc files. And WordStar for Windows, being a 16-bit Windows app, no longer runs under recent Windows OSes like Windows 7. The only reason I have any access to my WSWin files is via MS Word XP. Microsoft released a patch some time back that allowed Word ’97 thru ’03 to read WSWin files.

In short, I have bunch of lwp (WordPro) files that I can only access as long as I can still run WordPro and a bunch of wsd files that I can only access via MS Word XP with the WSWin service pack. I could go in with WordPro and Word XP and save each file one by one as a doc file and then change the date and time. (These programs don’t support odt.) Very cumbersome.

Maybe there’s some way to get LO Writer to open lwp and wsd files? Or some way to mass convert them to odt (preferred) or doc (second choice). Either would be a god-send. I only keep Lotus WordPro and MS Word XP installed on my computer for the purpose of accessing those files. LO Writer and SO TextMaker meet all my word processing needs.

The good news is TextMaker supports odt, so I can just set it to always save in that format.

sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
Tom_ZC wrote:Thanks to you
Interesting situation. I can see now why you’re worried about using a more permanent document format.

WSD

So the good thing is that it appears that at least the oldest format, WSD, is not a binary format. This means you can simply open it with any text editor or with MS Word or LibreOffice. The document won’t be imported per se, so there won’t be any formatting. But at least you’ll have your document text. There will probably be some weird characters used for formatting, but you should be able to use Find/Replace to get rid of them. Again, this could be scripted with a macro to run on all files in a folder.

Another option is to use this converter for WSW to text format. It doesn’t appear to support batch mode, though.

The best option appears to actually be StarOffice, which as you probably know, is an ancestor of OpenOffice / LibreOffice. According to this post, “The SO [StarOffice] import macro does a better job of retaining WS7 formatting than WSWin.” You can download the old version 7 of StarOffice for free for Linux or Windows here. With that, you can surely create a macro or possibly even use a batch convert method similar to the ones I mentioned in my previous post to convert all WSD files in a directory to a different gateway format. Then, you can use LibreOffice to batch convert from that gateway format to ODT.

LWP

This one is a lot easier. LibreOffice supports LWP import just fine, and you can use one of the batch conversion methods I linked to in my previous post to covert LWP to ODT.

Hope this helps, and I hope you can free up your important data.

___________________________________
LibreOffice 4.4, Ubuntu 15.04

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
LWP headless conversion
As mentioned, LO v4.0 can open LWP files natively, although I suppose that the one file extension may have through several file format versions. I’ve only done a rudimentary test using this LWP file I managed to locate on the internet. It appears to open OK, although without access to Lotus WordPro I have no idea what the original looks like. You can perform batch conversions of files by using LO in headless mode. I downloaded the example LWP I linked to and renamed it a.lwp for convenience. I opened it via the usual File > Open… method and saved it as convert_via_GUI.odt. I then made sure LO was not running and executed this command from a Linux terminal:
/opt/libreoffice4.0/program/soffice --headless --convert-to odt:"writer8" a.lwp

This produced a.odt which I renamed to convert_via_headless.odt. Both resultant ODTs are attached. I imagine they may well not be perfect in terms of formatting, but the text content seems intact. Notice that the two files are not identical. This is due to minor differences in things like the thumbnail icon, the system settings, and a layout_cache file, and the manifest XML file. The only difference in the content is the inclusion of <office:text text:use-soft-page-breaks="true"> in the GUI version versus <office:text><office:forms form:automatic-focus="false" form:apply-design-mode="false"/> in the headless version. You should be able to conduct much more effective tests at your end.

AttachmentSize
convert_via_GUI.odt 37.72 KB
convert_via_headless.odt 28.17 KB
sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
Thanks oweng. I’m curious
Thanks oweng.

I’m curious about that WSD format now myself. I tried to find one on the internet somewhere, but nothing turned up.

___________________________________
LibreOffice 4.4, Ubuntu 15.04

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
text, but not as you know it
WSD is (was) text, but full of control characters and dot commands. It is like SGML gone wrong. I know there is no such thing as “plain” text (it is always ASCII, or UTF-8, or whatever), but I must say this brings back some horrid file format memories and makes me thankful that these days I use text files under Linux for a lot of basic content. Long-term archival prospect is much better. Think of what DOC will look like in 20-30 years and it will make WSD look wonderful. Shock
sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
oweng wrote:Think of what DOC
oweng wrote:
Think of what DOC will look like in 20-30 years and it will make WSD look wonderful. Shock
No kidding. Binary format + Microsoft gibberish mixed in. Not an appealing prospect. Not to mention the fact that documents have become DRASTICALLY more complex in recent decades.

___________________________________
LibreOffice 4.4, Ubuntu 15.04

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
blast from the past
Yeah, you can say that again, particularly about the complexity. Last one before I pack it in for tonight. I managed to pull some install WS_ files, probably WordStar templates (WST), from a diskette image for WordStar for Windows v2.0 (US version).
$ ls -l DEFAULT.WS_ 
-rw-r--r-- 1 oweng users 8361 1994-03-01 01:00 DEFAULT.WS_
$ hexdump -n 128 -C DEFAULT.WS_ 
00000000  4b 57 41 4a 88 f0 27 d1  03 00 0e 00 00 00 33 12  |KWAJ..'.......3.|
00000010  20 26 37 58 68 78 78 77  71 01 43 64 44 86 07 99  | &7Xhxxwq.CdD...|
00000020  64 1c d2 53 9e 18 73 83  86 1c 60 71 80 38 2d 22  |d..S..s...`q.8-"|
00000030  65 25 62 75 dd 62 5d a8  64 b6 e0 b6 e0 9d a9 46  |e%bu.b].d......F|
00000040  09 4e 14 4f 59 96 25 16  52 52 39 de 48 38 39 3a  |.N.OY.%.RR9.H89:|
00000050  df 97 79 39 db 9d ed b8  8b 7e 6e 30 df 0e 30 e0  |..y9.....~n0..0.|
00000060  89 db 9d b8 3a e1 76 b0  dd 79 df 9e f8 eb 8e f0  |....:.v..y......|
00000070  df 97 76 0e 1b 7b bd ed  9e 2d d6 52 ed 21 6e 3b  |..v..{...-.R.!n;|
00000080
$ hexdump -n 128 -c DEFAULT.WS_ 
0000000   K   W   A   J 210   �   '   � 003  \0 016  \0  \0  \0   3 022
0000010       &   7   X   h   x   x   w   q 001   C   d   D 206  \a 231
0000020   d 034   �   S 236 030   s 203 206 034   `   q 200   8   -   "
0000030   e   %   b   u   �   b   ]   �   d   �   �   �   � 235   �   F
0000040  \t   N 024   O   Y 226   % 026   R   R   9   �   H   8   9   :
0000050   � 227   y   9   � 235   �   � 213   ~   n   0   � 016   0   
0000060 211   � 235   �   :   �   v   �   �   y   � 236   �   � 216   
0000070   � 227   v 016 033   {   �   � 236   -   �   R   �   !   n   ;
0000080

That is actually looking more like binary to me. Oh how I do not miss those days. LOL

EDIT: I forgot about these old underscore extension (*.WS_) files being compressed. Refer comment #12 below. Sample file is now attached for others.

AttachmentSize
WSW2_us_DEFAULT_template_compressed.zip 7.63 KB
sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
Ugh that’s not good. Well,
Ugh that’s not good. Well, for Tom_ZC, I would say that if he can’t read anything by opening the WSD files with a text editor, then he should try StarOffice 7 to convert them.

___________________________________
LibreOffice 4.4, Ubuntu 15.04

Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
Thanks
Many thanks to you both for your info. Now I’m armed with what I need to convert my older more obscure files over.

One more thing about file types: I’ve avoided Microsoft’s new docx file format like being photographed in the shower by Jodi Arias. I have a pen pal in France. She corrects my French and I correct her English. She always sends me docx files. I won’t even save them to my hard drive. I’ve been saving them to a flash drive and then converting them to doc files before saving them over to the hard drive. The docx format and the Ribbon are the two big reasons why I dumped MS Word for LO Writer.

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
KWAJ ... where do I know that from?
My comment #9 is partly wrong. I had a funny feeling I had seen the leading “KWAJ” signature before and indeed it (plus the underscore at the end of the file name) indicates the file I obtained was a compressed version of the final DEFAULT.WST template. Naturally it will look binary. I will go back and attach the file for the interests of others.

I managed to find an uncompressed equivalent STANDARD.WST template, from the German version of WordStar for Windows 2.0 and it reveals a clearer picture:

$ ls -l STANDARD.WST 
-rw-r--r-- 1 oweng users 33801 1994-03-17 18:00 STANDARD.WST
$ hexdump -n 128 -C STANDARD.WST 
00000000  23 20 56 56 20 23 20 35  2e 30 30 0d 0a f1 01 48  |# VV # 5.00....H|
00000010  00 0a 00 10 00 64 00 01  00 01 00 00 00 00 00 00  |.....d..........|
00000020  00 00 00 00 00 00 5a 00  5a 00 ff ff 00 00 01 00  |......Z.Z.......|
00000030  03 00 01 00 00 00 02 00  03 00 00 00 00 00 02 00  |................|
00000040  00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00000050  00 00 00 00 00 00 00 00  01 00 3b 00 00 00 00 00  |..........;.....|
00000060  00 00 00 00 0f 00 04 00  05 00 02 00 00 04 e8 03  |................|
00000070  0c 00 02 00 04 00 03 00  07 00 08 00 09 00 00 00  |................|
00000080
$ hexdump -n 128 -c STANDARD.WST 
0000000   #       V   V       #       5   .   0   0  \r  \n   � 001   H
0000010  \0  \n  \0 020  \0   d  \0 001  \0 001  \0  \0  \0  \0  \0  \0
0000020  \0  \0  \0  \0  \0  \0   Z  \0   Z  \0   �   �  \0  \0 001  \0
0000030 003  \0 001  \0  \0  \0 002  \0 003  \0  \0  \0  \0  \0 002  \0
0000040  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0  \0
0000050  \0  \0  \0  \0  \0  \0  \0  \0 001  \0   ;  \0  \0  \0  \0  \0
0000060  \0  \0  \0  \0 017  \0 004  \0 005  \0 002  \0  \0 004   � 003
0000070  \f  \0 002  \0 004  \0 003  \0  \a  \0  \b  \0  \t  \0  \0  \0
0000080

I will attach this file to this post for others.

AttachmentSize
WSW2_de_STANDARD_template.zip 5.24 KB
Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
I did find one, and only one,
I did find one, and only one, reason to use doc over odt. The brainstorming/mind mapping software Xmind (http://en.wikipedia.org/wiki/XMIND) will export to a doc file, but not to an odt. This disappoints me because Xmind is open source and free software like LibreOffice. I guess I could export to a doc file and then use LO Writer to convert the file to an odt file. It’s an extra step, but maybe it’s worth it since odt is so much better. I think I’ll still go with odt. I’m planning to write to the developers of Xmind requesting that they support an export filter to odt. If anyone else here would like to, please urge Xmind to support open source by supporting odt.

Here’s another interesting tidbit. I set Writer and Calc back to automatically save in their native formats and I converted all the files for my novel from doc to odt. When I double clicked on one the odt files in the standard Windows file explorer, it automatically pulled the file up in LO Writer. However, when I double clicked on the same file with the Magellan File Explorer, it showed me a bunch of xml files and directories. If I double clicked on one of the xml files, it would pull it up in Internet Explorer (which I almost never use and is not my default browser) even though xml files are associated to XML Editor on my PC.

And here’s an annoying tidbit. I temporarily had MS Office 2010 trial on my PC in order to help a friend who’s a novice computer user. The thing hijacked my file associations! I had doc, docx, and odt files associated to LO Writer, and MS Office hijacked every one of them from my using Word 2010 ONE TIME! I couldn’t believe it. There was no message to the effect of, “Do you want Word 2010 to be the default program for this?”

Edit: Another update for you. Today I opened in LO Writer one of my plays that I wrote in Lotus WordPro (and saved as an lwp file). It opened with all the content in fact and with only a few minor formatting issues. I was able to quickly fix the formatting and save the document as an odt file. Success! Thank you for your help.

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
FreeMind
Have you seen FreeMind ? It can import/export to a lot of different formats.

The other matter about file associations is unsurprising. MS own your system. That is merely a gentle reminder Smile Magellan File Explorer I have not idea about. It sounds like it is treating the ODT as a ZIP file, which in some contexts is unsurprising.

PeterRoots
Offline
Last seen: 2 years 16 weeks ago
Title: ★★★★★
Joined: 25 Oct 2011
Posts: 941
I rather like FreeMind but,
I rather like FreeMind but, just in case you don’t like java apps then https://sourceforge.net/projects/vym/ VYM also looks good. It reads FreeMind files and can export to odp as well as a lot of other things
Linux Mint 15 LibreOffice 4.1.1.2 : OpenSUSE 12.3 LibreOffice 4.0.3
oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
more info on FreeMind / Freeplane
Thanks for the VYM link. Any recommendation about it i.e., are they looking to improve ODF support / inclusion? I have not used VYM and it has been some time since I used FreeMind. My general trend away from Java also means that I will not likely be going back to FreeMind.

For those that are interested Freeplane is evidently a fork from FreeMind that may also be worth considering. It still appears to be Java-based, but more community focussed and the code has supposedly undergone considerable review and change. More importantly there is clear documentation on how they are considering making use of the packaging aspect of ODF v1.2 (detailed here and here on the file format wiki page) to directly include Freeplane mind maps. How successful this will ultimately be appears to be a work in progress.

PeterRoots
Offline
Last seen: 2 years 16 weeks ago
Title: ★★★★★
Joined: 25 Oct 2011
Posts: 941
I only tried VYM out to reply
I only tried VYM out to reply to this thread, so not much experiance of it! Not being Java based it looks nicer than FreeMind (or Freeplane I just tried out of few minutes ago). I prefer the FreeMind map appearance over VYM though. FreeMind produces odt vs VYM odp but both are only essentially outlines based on the map rather than the map themselves. The other export options are more useful though, in both (only sampled a few). I also tried installing Braindump (Part of Callligra) which has no help, no option to open or save a file and seems to do practically nothing useful at all Smile For usability I would go for FreeMind or FreePlane despite the clunky Java look. I used FreeMind quite a bit a long while ago but have had not need for it recently
Linux Mint 15 LibreOffice 4.1.1.2 : OpenSUSE 12.3 LibreOffice 4.0.3
priehl
Offline
Last seen: 2 years 39 weeks ago
Title:
Joined: 19 Mar 2013
Posts: 7
sb73542 wrote:I agree with
sb73542 wrote:
I agree with Oweng. A huge advantage of the ODT format is that ODT is basically like a ZIP archive that contains, among other things, the text portion of the document in XML format. So if for some reason all version of LibreOffice and OpenOffice.org all mysteriously disappeared from the universe, at least you could unzip them with almost any archiving program and read the XML file.

Excellent to know. However, I don’t see a way to access it on Mac – Windows is no problem after changing the suffix to zip.

oweng
oweng's picture
Offline
Last seen: 4 hours 37 min ago
Title: ==Moderator==
Joined: 26 Jan 2012
Posts: 3064
MacOS options
You need something like StuffIt Expander or the free Zipeg on MacOS. I’ve used the former but not the latter.
priehl
Offline
Last seen: 2 years 39 weeks ago
Title:
Joined: 19 Mar 2013
Posts: 7
Ironically, I just removed
Ironically, I just removed Stuffit from my Mac a few days ago. It works, indeed; I had no success finding the xml text in Zipeg for some reason.
Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
oweng wrote:Have you seen

oweng wrote:
Have you seen FreeMind ? It can import/export to a lot of different formats.

The other matter about file associations is unsurprising. MS own your system. That is merely a gentle reminder Smile Magellan File Explorer I have not idea about. It sounds like it is treating the ODT as a ZIP file, which in some contexts is unsurprising.

Yes, I’ve also considered FreeMind. Since LibreOffice doesn’t have an equivalent to MS OneNote, I’ve been considering the free note taking software talked about on this Wikipedia page:
http://en.wikipedia.org/wiki/Category:Free_notetaking_software

I only began using Xmind a few weeks ago for planning out a new novel. I might check out FreeMind or ViewYourMind. I used to write all that stuff out by hand on large paper.

oweng wrote:
The other matter about file associations is unsurprising. MS own your system. ….

Which is yet ANOTHER reason why I’m considering getting rid of Microsoft altogether and going with a Linux PC. I used to use Ubuntu on an old clunky machine for certain things. I loved it. I won’t give you the laundry list on why I’d like to jettison MS completely or I’d have to write a dissertation. The only thing keeping me using a Microsoft operating system is software that I use, which I’m not sure if I could get working under WINE. There’s the Final Draft stageplay/screenwriting word processor and then there’s Canon’s digital photography software ZB Module. Some of my photography software WILL run under WINE such as Photoshop and Paint Shop Pro. Another one I’m not sure of is Ulead/Corel PhotoImpact. For photography there’s strong incentive to use either Windows or Mac. But there might be a way to make it happen with Linux.

For writing my books, there’s no reason on this earth to stick with MS. LO Writer handles them just fine without subjecting me to the atrocious Ribbon interface.

sb73542
Offline
Last seen: 1 day 22 hours ago
Title: ==Admin==
Joined: 28 Sep 2010
Posts: 167
priehl wrote:sb73542 wrote:I
priehl wrote:
sb73542 wrote:
I agree with Oweng. A huge advantage of the ODT format is that ODT is basically like a ZIP archive that contains, among other things, the text portion of the document in XML format. So if for some reason all version of LibreOffice and OpenOffice.org all mysteriously disappeared from the universe, at least you could unzip them with almost any archiving program and read the XML file.

Excellent to know. However, I don’t see a way to access it on Mac – Windows is no problem after changing the suffix to zip.

Possibly one of the links for Mac down toward the bottom of this page will work:
http://www.7-zip.org/download.html

___________________________________
LibreOffice 4.4, Ubuntu 15.04

Tom_ZC
Offline
Last seen: 1 year 20 weeks ago
Title: ★★★★
Joined: 15 Apr 2013
Posts: 67
An update: Turns out Xmind is
An update: Turns out Xmind is NOT free GNU software. I went to use one of it’s features and got one of those annoying messages to the effect of, “This feature is available in the paid version only.” I felt deceived, but turns out Xmind’s web page doesn’t make any claims about it being free GNU. My confusing it for free came from this Wikipedia page, which incorrectly lists Xmind as free note taking software:

http://en.wikipedia.org/wiki/Category:Free_notetaking_software

I guess I could pay for the thing or use it with limited features. Or I could give FreeMind a try.

livebe
Offline
Last seen: 1 year 11 weeks ago
Title:
Joined: 11 Mar 2013
Posts: 4
Tom, I would never recommend
bq.. Tom, I would never recommend DOC over ODT as the former is a binary format. When a binary format file gets corrupted, you throw it out and start again 99 times out of 100. At least with a ZIP file containing XML (i.e., ODT) you are more likely able to recover pieces. The other problem with using any non-native format is that you are constantly at risk of the feature being deprecated or impaired in some way. Even in LO features can get turned off with each release e.g., v4.0, v4.1, etc.

Well, I must dis-proved that.

Here it is:
1/ Take some odt file and make a copy of it
2/ Open the copy into some text editor(e.g. Notepad++) and randomly update/insert or delete a SINGLE char, then save and close the file.
3/ Try to open it in LO 4.3.4 it fails – LO will try to open it, notices it has been corrupted, click YES for attempting to fix that and you are thrown with a general error.
Close LO
4/ Rename you copied odt as a zip file and try to open it with say 7-zip, you’ll see it won’t.
This is consistent with error recovery in zip and other archiving formats.

col48a
Offline
Last seen: 1 year 5 days ago
Title: ★★
Joined: 20 Feb 2015
Posts: 11
Wait a mo
oweng’s claim in post #2 doesn’t actually claim that the recovery could do more than get back the text. And this would be by using gedit, or Notepad or something similar in the first phase rather than LO.

As you imply, even the smallest change to a zip file will disturb the error recovery, possibly beyond repair.

I think the point is that using a binary format for a document removes the opportunity for text recovery by a simple method. If recovery is possible at all, it has to exploit the way the file is structured – ie it’s clever software – and the damage can easily be too much for success.

Ubuntu 12.04 64-bit.
floris v
floris v's picture
Offline
Last seen: 15 hours 47 min ago
Title: ★★★★★
Joined: 16 Nov 2014
Posts: 289
Correct. Even if recovering
Correct. Even if recovering your text from a damaged binary file is a pain, at least you can recover it. You can recover nothing from an irreparably damaged zip archive. Add to that that if you have a lot of embedded images and the file gets truncated, and that can happen, the content.xml file will almost surely get lost, and that’s the most important part of the entire archive. That’s another reason to insert your graphics as links.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.