Code Cleanup Culls LibreOffice Cruft
mikejuk writes with an interesting look at what coders can get around to after a few years of creating a free office suite: dealing with many thousands of lines of deprecated code: "Thanks to the efforts of its volunteer taskforce, over half the unused code in LibreOffice has been removed over the past six months. It's good to see this clean-up operation but it does raise questions about the amount of dead code lurking out there in the wild. The scale of the dead code in LibreOffice is shocking, and it probably isn't because the code base is especially bad. Can you imagine this in any other engineering discipline? Oh yes, we built the bridge but there are a few hundred unnecessary iron girders that we forgot to remove... Oh yes, we implemented the new chip but that area over there is just a few thousand transistors we no longer use... and so on." Well, that last one doesn't sound too surprising at all. Exciting to think that LibreOffice (which has worked well for me over the past several years, including under the OpenOffice.org name) has quite so much room for improvement.
There are probably dozens of extra nails that were just hammered in rather than be removed. There are extraneous pieces of lumber.
And a house that was remodeled? I've seen newspaper used as filler. I've seen layers of roofing, with things buried in between layers.
Frankly I don't know what's inside my walls, and I'm not sure I want to know.
Nice alliteration.
It would not be very surprising to see a lot of dead code.
I maintain the code for MoreTerra, a Terraria map editor program and I'm pretty sure I've got dead code in there and that's a pretty small project.
With a large number of people working on the code it likely ends up slowly clogging up as no one quite knows what the others are doing.
Dare I ask what type of dead code exists in something extra huge, but closed source, like the Windows code base or for MS Office? But I'd
bet for all MS's faults that the code for Norton Antivirus is 10x worse.
Sounds like they already put a lot of work into this, but someone should tell them that you can automate things like removing unused code.
Troll is not a replacement for I disagree.
No one really cares about unused functions, it doesn't degrade the resulting binary at all if the compiler you are using is any good. It may help in readability though. Refactoring the code is a whole other thing, but it don't think that is what they are doing here.
I tried libreoffice a few days ago and as it repeatedly crashed within minutes of work, I reinstalled openoffice. For now. It looked promising and was significantly faster.
cb
I've been working on a project where there were 3 separate wrappers around a database, each returning different objects containing the same data... So you had to convert those each time two modules using different wrappers needed to communicate. I tried to clean it up a bit, but eventually I stopped because my manager was frowning upon that because "I broke working code". Also there were parts that I didn't know if they were still in use. I also ran a profiler and found 80% of the functions never got called. That doesn't mean it's dead code of course, but looking at the function names I got an eerie feeling with a bunch of them. Anyway, I learned a lot about how not to manage software, I quit the company since then and I can only hope things have changed over there.
"It's too bad that stupidity isn't painful." - Anton LaVey
Get over it.
...lots of stuff is left lying about which might not be used any longer on the off chance that it might be adapted to some future purpose. Sounds like genetics.
so most people wont notice a new build.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
Can you imagine this in any other engineering discipline? Oh yes, we built the bridge but there are a few hundred unnecessary iron girders that we forgot to remove...
Those would be perfectly valid if upon discovering your girder was 3 inches too short you could instantly create a copy of it, set the original aside, then alter and test that copy of the girder. Then you might leave a few extras lying around.
I never get crashes with LibreOffice. Whenever I try Word on some documents (docx) I get a crash. I was completely unable to edit some documents in Word (sent to me by colleagues) until I opened them in LibreOffice, saved them in doc format, then reopened them in Word. It happens with distressing regularity. I find LibreOffice much more stable than Word personally. The worst part is when once I edited a doc in Word, saved it, and when later tried to open it again had a similar problem. I am not sure what document elements cause this but it's a sad state of affairs when LibreOffice is not only more stable (for me), but handles better MS own file format (even though there are still big deficiencies in the docx file handling in LibreOffice). So, stability issues? I guess it depends on your computer.
I like my dinosaurs feathery, and my pterosaurs hairy (or is it pycnofibery?)
Bridges often have unused structural elements: walk-ways made unsafe by modern traffic levels, maintenance accesses unused for safety reasons, supports made redundant beyond the factor of safety by bridge improvements, etc. Chips and boards too: FPGAs with 10% utilization, chip designs re-purposed with functional components disabled, subsystems replaced in boards by new designers not confident enough to remove the old design, etc.
Cruft in software is more often removed because (1) software has a potentially longer lifetime than hardware and (2) it's a lot easier to remove an uncalled function from a program than a girder from a bridge! Software cleanup should be an expected and planned part of a project's life cycle.
Mephisto is a German demon
LIbreOffice hasn't been OO in well over a year. But nice try with the trolling.
Oh yes, we built the bridge but there are a few hundred unnecessary iron girders that we forgot to remove...
Well, look at bridges built in the 1800's compared to the ones today. Would we build a modern bridge today using wrought iron links http://en.wikipedia.org/wiki/Clifton_Suspension_Bridge? Each building made in a certain period in a way represents a degree of refinement compared to its predecessors. Better materials, better methods. Buildings in general cannot be "cleaned up" the way code can, where "cruft" today was yesterday's conservative design.
Read a book about the differences in the construction of the World Trade Centers versus the Empire State Building, for example (the WTC has sibling buildings still around using the same techniques, such as the Aon [nee Amoco] Building in Chicago)...
The issue is similar to the Japanese concept of "heijunka." http://en.wikipedia.org/wiki/Production_leveling
Pressure for speed of production of software vs. quality of product in speed and lack of bugs demand compromise. No one will use your software if you get beat to market by the competition. No one will buy your product if it's unreliable.
Pay no attention to the man behind the curtain with all your metadata.
The quoted section in the summary asks if we could imagine this in other engineering disciplines. As the rest of the summary points out, it happens all the time in microchips. It also happens a lot in civil engineering, including bridge building. Removing things takes work. Unless there's work to be saved by doing it, or some way to profit from selling what's removed as scrap or it's a safety issue to leave it most engineers won't remove old parts of a structure. Consider underground pipes. How often are they removed when they're replaced? If the new ones are being laid down where the old ones went, they'll be replaced. Otherwise, 90% of the time they'll just leave the old ones there. Same goes for just about everything. Old installations of any kind are full of stuff that no longer serves any purpose. Brackets and supports for heavy equipment that isn't used anymore, old wiring and panels, concrete slabs that something mystery object used to sit on, etc. When was the last time you saw anyone take away some 30 ton piece of equipment then pay more money to have the floor where it used to sit un-reinforced? Now, sometimes they do. Usually it's when the place is being sold and the new owners are re-modelling. Other times the owners do decide to do a major cleanup. That's exactly what's being done here with libreoffice. Makes it no different than any other engineering discipline then.
Incidentally, if it's truly "dead" code, then it shouldn't actually be compiled, so it's not like the bridge engineer left in a bunch of extra girders, it's more like he's keeping addendum 6-c to revision 12b of the plans for section 3 in the same file cabinet as revision 13 rather than shifting it to a storage box and warehousing it.
If you removed all of the dead code from AOL you'd have nothing left...
I used to work at a large, evil company that did a complete review every release of every single line of code in their products, requiring at least two and sometimes four (for critical areas) devs to sign off on it. Let me tell you; there was _no_ dead code.
Human DNA (and just about every other species as well) is full of things like inactive duplicate genes (some with slight alterations), pieces of old retroviruses, and other mutations and replication errors that have been "commented out". Plus a whole lot of sequences which we don't know what they're good for yet.
I've seen chemical plants built with millions of dollars worth of unnecessary piping and valves, because the project timeframe meant that it was cheaper to install extra connections that might never be used and save engineering time than waste time re-engineering it.
If removing unnecessary items can save thirty thousand dollars (say) at the cost of three days, removing the cruft is only worth it if the delay costs less than ten thousand a day.
Stop opening Word documents with non-MS software and you won't have any problem. Other software are more likely the reason of incompatibilities. I use MS-Word every day and I have not seen such problems since Office 2003.
You know the Germans always make good stuff.
Orders of magnitude :(
A name change doesn't make it any more stable. It's still the same old codebase.
Exciting to think that LibreOffice has quite so much room for improvement
Sorry but removing the dead code is not really "room for improvement". It's rather "fix a code that lack(ed) a proper project management".
Slashdot, fix the reply notifications... You won't get away with it...
Well this completely explains my confusion over why the latest release of LibreOffice was smaller than the older version.
LibreOffice/OpenOffice.org Writer has usually been good at handling traditional .doc files. (OO got much better when early 2.x versions came out) I'm usually able to share these documents with Word users and make changes without any major issues.
Docx is a whole different story. I recently opened a moderately complicated docx in OO Writer. This document had one of those auto-generated table of contents that corresponded to section headings, several embedded pictures, a few tables, and lots of lists. While Writer would let me see what was in the document, the editing capability was completely shot to hell. Lists were broken, the table of contents was a read-only object that Writer wouldn't (or couldn't) let me change, style/formatting was glitchy in general, and track changes was broken. Converting that document to another format just made things worse. Don't even think of using OO/LO with Docx files unless the only thing you have to do is read them.
"It is a denial of justice not to stretch out a helping hand to the fallen; that is the common right of humanity."
Except building a house is a one time endeavor. A better analogy would be to build a house on the foundation of another house and then remodel it a dozen times and then add on a few additions. Soon there after divorce your spouse and find another to guide the new additions and find another site upon which you can move that house to and hopefully not have to deal with that awful foundation.
Intel's microcode to support 16-bit protected mode became obsolete as soon as the 80386 was released, but they had to support it for backward compatibility, in case someone tried to install Windows 3.0 on an IBM AT clone, for instance. Probably that microcode has been carried forward ever since. Also, there are a lot of CISC instructions such as SCAS* with the "REP" prefix which were heavily used in assembly language in the eighties, but which are now deprecated and typically slower than the RISC-style replacements.
If it is a house that I have owned, it would be Pez dispensers. Whenever we do remodeling, we make a point to slip a pez despenser into the walls. My wife and I figure that some day a long time in the future, someone will have a mildly amusing story.
And then rebuild it from scratch with new age materials.
Of course, that behavior is quite similar to what happens when you open the moderately complex docx document in Word.
Faster! Faster! Faster would be better!
Dont understand... Dont you ave one with a car?..
-no sig today-
try resizing the main window using the upper left corner...
This guy should be fired. Look at his productivity metrics results, Negative!
but recent (3.x) versions of OpenOffice ate my kids documents. It really sucked. From what I can gather it's a known bug in the document recovery module that hasn't been fixed to this day. The program crashes, writes a blank document out as the 'recover' document, then cheerfully overwrites all your original file and any of the automatically made backups. I suppose that somewhere along the line there was some user error. My kid probably could have said 'no' to something and stopped the whole mess. But seriously, she shouldn't have too. I've got a 500 frickin' gig drive in her machine. The biggest word doc I've ever seen in my life was 5 megs (mostly pictures). Why the hell do we still delete shit? Just make a huge undo buffer or something. I've got half a fscking terabyte. Come on OO.org, just use it already!
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
If you have newspaper or other similar material in your walls, which wasn't processed and designed as insulting filler, I have one word for you: mold.
You'd better know.
Actually I've found LO/MS Office to be pretty much mutually exclusive when it comes to any complexity of formatting. This is why i give it to home users that don't have kids or who just has young kids but not those with HS or college kids because if the teacher is using MS Office anything they give to you will be word salad in LO and anything you give to them will be word salad in MS office. And why does so many LO users find that hard to accept? they are trying to reverse engineer a binary format without jack shit for source code to look at, the fact they can open up as many as they can is a miracle and they should be applauded for it. But you take a complex doc with headers, footers, tables, etc and while i've had no problem opening that in any version of MS Office I have from 2k-2K7 that same doc will take a big old shit in LO. by the same token i make the same doc in LO with the same headers, footers, and tables and MS office will either crash or output gibberish. Does that make one better than the other? noooo, it just makes them incompatible.
That is why I give away LO to home users like I said but don't even mention it to SOHOs and SMBs because if you are having to share with MS Office users it can quickly become a mess. So I'd say it is just using the right tool for the job, LO when its home users who'll mainly be printing or not sharing with businesses, MS office for business and HS or above education.
ACs don't waste your time replying, your posts are never seen by me.
No, a better analogy is to build a house (full of extra materials as the parent said), and then use a giant replicator machine to mass-produce the house, almost instantly, and create thousands and thousands of new homes using that house as the basis. The wasted material in the one house is bad, but not that bad because it's one house, and it takes extra time and labor to do it more efficiently. But multiplied across thousands of identical copies, that wasted material adds up a lot. Plus, it's inefficient and you could have a better-performing house by doing a better job with the small details (better at energy efficiency for example). The slight increase in energy efficiency with that one house, realized by spending a bunch of extra time and effort removing wasted materials and doing a better job with various small details (like making sure the house wrap is applied extremely well rather being hurried and missing some staples in important places), won't amount to much with just the one house. However, multiplied across many thousands of houses, those energy savings add up to a lot.
The fact that software is easily and quickly replicated with perfect precision and little or no effort or time really makes it hard to make good analogies for it without resorting to Star Trek-style replicators; it's the only technology we have that's like that. And because it can be and is copied so easily, very different dynamics apply to it than to many other fields of endeavor.
If you have newspaper or other similar material in your walls, which wasn't processed and designed as insulting filler
What material in your walls could be more insulting than newspaper?
A lot of European cities were rebuilt on the same place over and over again, after wars, natural catastrophes, crazy rulers,...
There are layers and layers of building foundations and walls under the modern (and especially not so modern) buildings. It's not uncommon constructions to be stopped for excavations to take place.
In love, war and slashdot discussions, everything is allowed.
Not always. Aren't the Germans huge fans of homeopathy?
Some of them also have very questionable taste in style. Just look at some of their wacky-looking cars; compare a top-of-the-line Porsche to any Ferrari, for instance. The Italians have been the masters of automobiles as art for quite some time, but Porsches and BMWs usually have very bad proportions. BMW's sedans look good, but every time they try to make a sports car it just looks weird, and they do much better when they stick to making the engine and supplying it to someone else, like with the McLaren F1. A lot of VWs have rather questionable styling too. But the Mercedes cars are a giant exception to all this; those are some of the best-looking cars out there, short of the Ferraris. I wonder if the Germans styling those cars come from a different region of the country, or if they outsource their design.
"WORKSFORME"
The most unhelpful response to a bug report ever.
But very helpful response on trolling.
In love, war and slashdot discussions, everything is allowed.
Well, there are a variety of fiber, composite and other materials with higher R-factors per volume, but that's beside the point.
Newspaper as in processed recycled newspaper bought as insulation from Home Depot or the equivalent is one thing. Stuffing newspapers into your wall after receiving them in the post and reading them (a common practice) is quite another. The latter retains moisture and can lead to mold, rotting and structural damage, just to start with the most obvious problems.
Why the analogy of bridge building - they are not even close.
If I want to build a bridge I would use a plan, that someone already created and guess what the bridge would not change afterward. If the bridge did not meet your needs you would build a new one.
Building a bridge is like installing Windows XP, in both cases the work has been done already by a previous person. You are just coping it.
http://www.developerdotstar.com/mag/articles/reeves_design_main.html
http://blogs.msdn.com/b/steverowe/archive/2005/02/28/381910.aspx
Unused memory is wasted memory. Windows 7 intentionally uses as much memory as possible if it's available rather than paging during application use. I wouldn't be surprised if competing operating systems like OS X behaved similarly. If something ends up needing that memory, the operating system will happily give it away. It's silly that your argument constantly reappears in various online forums, because it signifies a lack of knowledge about how modern memory management actually works.
Something tells me you're not the only one.
"Wooooosh"
I see your anecdote and raise you another - I use recent editions of MS-Word every day too, and often see exactly the problem the GP post is alluding to. I don't open Word documents with non-MS software, and nor do my team members.
I usually ask the sender of the document to send it through again in another format. I hadn't thought of using LibreOffice to open the offending file instead. Thanks GP :)
NO-body expects the Spa... oh, bugger.
Building a house isn't a one-time endeavor. Much like code, houses are never 100% finished. They're frequently repaired and less frequently remodeled, renovated, or expanded. If you look at photographs of the same house over the span of a century or more, it's sometimes hard to believe that the final version is the same building as the original. And when people work on their houses, they usually go for the most cost effective approach, even if that means leaving no longer used stuff in place because it would be more expensive to rip it out. Look inside the wall of an old house, and you'll be amazed at the stuff you find.
There's no point in questioning authority if you aren't going to listen to the answers.
The point of TFA is that ... no ... it's not the same codebase anymore. Not by a long shot.
Put identity in the browser.
LIbreOffice hasn't been OO in well over a year. But nice try with the trolling.
It also includes a fair bit of Go-oo. That and the work to remove Java means it's a substantially different code base.
It still has a ways to go before it can do what OpenOffice.org can do - and now that OpenOffice.org is under Apache I expect OO will continue to improve (also).
If people going to mod they should do so fairly. This is no less flamebait than it's parent. modding isn't supposed to be a way of picking a side.
That's nothing. Whoever built the wall that exists between my office and the dining room left inside a leather dog collar and a half-dozen pork (I think) rib bones.
We've also found a cast iron floor lamp inside of a wall, as well as several hundred copies of the Saturday Evening Post that are positively impossible to drill through.
Pez dispensers seem so...basic.
Kid-proof tablet..
Yeah, like SAP.
Oh wait...
Also LibreOffice often seems to crash when editing docx formatted documents, however recent versions of MS Office open odt documents quite well so I find it easier to save everything it odt.
null
Whilst I agree with you on the docx side, MS Office seems to work with odt documents quite well unless you use complex styles in OO (and most users won't/shouldn't) in which case you have to run a document recover when you open it in MS Office, but usually the output is still fine just with some missing features.
null
The difference with a house is that material is expensive and energy lost through poor insulation or construction is expensive. Bandwidth and memory are both pretty cheap, and with computers wasting memory tends to result in slower operation which is much harder to measure in monetary terms.
Consider this though. Would we have pretty high quality and very complex office software by now if we were not able to be a bit wasteful? Hard to say but I'd guess not, at least not for free. Still, I very much welcome code cleanup, wish I had the time to help.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
"...and it probably isn't because the code base is especially bad."
Yes, it does mean that. Thousands of lines of deprecated code implies it was written in a sloppy and disorganized way and this is not only an indicator but, I would argue, a definition, of a terrible code base.
Mercedes is from Stuttgart, south of Germany, closer to Italy. Not that I believe it matters.
Terrible excuse for MS software. Programs are expected to accept any kind of input properly.
I find your reply insulting.
Actually, building materials (at least here in the USA) are dirt cheap. Labor by far is the biggest factor in a house's cost. There are some exceptions, like appliances, cabinets, granite countertops, plumbing fixtures, etc., but things like framing, drywall, nails and other fasteners, even electrical outlets (except the GFCI ones) are all dirt cheap. That's why there's waste; the contractor doesn't care if his workers waste $2 worth of materials as long as they're fast. However, when the plumbers are putting in expensive polished faucets, they're probably more careful since replacing one of those is worth several hours of labor.
Something matters, because there's (IMO) an enormous difference between the styling of Mercedes cars and BMW cars; it could just be company culture, but it could also be regional differences. After all, not that far away is France, home to several automakers, and I don't think anyone will deny that French cars have always had a very different design aesthetic from any German makes. Furthermore, there are cultural and linguistic differences between different parts of Germany. I'm just wondering if that's a factor in the differences between the various German automakers.
The factory where I work which must be over 60 years old, there are redundant cables and pipes everywhere. It would cost far too much money to figure out what is redundant and then remove it without disrupting anything else.
We remodelled our bathroom, and when removing the ancient bath we discovered two sealed packs of cigarettes from the 1920s.
Sold them to a collector for £3000 each.
Funny what you find, isn't it?
> pork (I hope) rib bones
FTFY
well, at least Pez dispenser is not a dirty diaper.
You can't handle the truth.
When the local city gets bus-sheds, while having to put up with commercials. There are many situations where you are given something for "free" and have to put up with extra stuff just in addition to what you wanted. I hear there are actually people who watch commercial breaks in order to see things they want on their television. Nothing is free, and you pay one way or another.
But is Libreoffice that bloated? The unused code might as well have been unused features. Compare the install-base of M$ Office and LibreOffice and the relative size of the excess becomes less relevant.
That being said, there are some things that OSS can do better. One thing is that there should be groups that tests their GUI. Another is groups that document features properly. The documentation of OSS has gotten a lot better since Youtube came along. Now, people are posting guides to software fairly often. But, there is still need for a coder community that includes other groups better. Perhaps even get some graphic talents along for the ride? Perhaps somebody who knows this stuff should do a Youtube vid about OSS project management?
QA? ...
Holy fuck. There's light.
Deleted
Unless you're talking about removing inefficient structures in frameworks, I fail to see how that would decrease build times. That said, I haven't developed land before but I presume there'd be a frightening amount of decrepit function in some of Europe's catacombs.
Now image this being a nuclair plant. On one hand one does want extra high engineering conditions applied. An extra pipe will cost a lot because of this.
On the other hand, removing cruft might cost very much. It might be low level radioactive, which makes it more expensive to get rid of.
You've probably got a dead cat. Sometimes they get stuck between the layers. In old houses they were sometimes put there intentionally as protection against witches.
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Newspaper is actually pretty handy - if you need to fill a hole in a hollow wall, you could end up pushing a kilo or two of filler in there before it starts to build up, but if you just stuff it full of newspaper first, you only need as much as you would to cover the surface. It might seem like a bodge, but really it just saves you materials. It can also be used to stop the draft from under the floorboards underneath a carpet. I've worked on houses that get wrecked by students every single year, and to be perfectly frank, if you have to bodge something up to make it look clean, it doesn't matter, because next year it will invariably be trashed again.
It is always tempting to say others' code is "cruft" and unused. Removing features and marking them "obsolete" is oh so innovative and _daring_!
But the reality is dull. OO/LibreOffice severely lag behind the MS version in performance (I'm running OO 3.3 and MS Office 2003). When I'm loading even a normal-sized Excel workbook (20 to 30 sheets with a lot of calculations in them), OO hangs... MSO doesn't.
Same for long Word documents. I use OO to write memos, but MSO is for longer works (unless I fancy to use LyX/LaTeX for the purpose).
The morale of the story is anybody can delete somebody else's code, but it takes real talent to improve the complexity of an algorithm. Microsoft at least used to set the right priorities in development (until a decade ago, I reckon :)
No, regional differences on that scale do not affect car companies that operate globally. These companies would be perfectly able to copy, say, American car designs, but they just won't do it. Audi, BMW and Mercedes are rather treated as brands by the respective car designers, which means that the cars must each have certain recognizable elements and an overall design that is related to other past and current cars that they built. Plus, different styles cater to different people. So it's both diversification and brand recognition at work there.
Good analogy.
But Libreoffice still uses Java. I don't see that fitting into your analogy, because the Java dependency really has to be removed. It was put there only because OpenOffice was in the hands of Sun. Now Java is in the hands of Oracle. The Java dependency has to go.
Yes - when I bought my last car, I fitted a new stereo. When I fitted the speakers, inside the doors were three old speakers, loads of wire, some beer cans, a screwdriver and a broken pair of pliers. I did a thorough search of the rest of the car and found a lot of spare wires and coins under the rear seat, but no cocaine
Sent from my ASR33 using ASCII
Wait; so they've had half their people working for half a year to remove code which isn't used anymore?
(Disclaimer here; I'm an occasional, hobbyist programmer at best. It's entirely possible that I'm missing something here, and if I am do please enlighten me.)
I wonder why this couldn't be automated. You make a program that runs through and makes a big list of every function in the source, deletes functions that aren't called anywhere, repeat a few times to deal with chains of unused functions, and you're done. It seems like exactly the kind of task a computer is designed to do. Have a few flags to tell it whether to treat commented-out functions calls as valid so you don't wind up removing the alternate version of routines while trying out experimental new versions, and whether to actively delete the functions or just feed them back to the programmers to examine themselves, and you're finished. If you really want to be clever, have it look for calls that technically exist, but due to the logic involved would never get invoked in any circumstances.
I'm not even sure it'd be fair to say it would take too long to develop such a tool. After all, once it's made for a particular language, then it's done; everyone can take advantage of it, and in a few years time when you decide you need to do another grand cleanup, no need to take up six months and half your team for the task.
Now don't get me wrong here; code optimisation is a different beast, and there's far more to maintaining a tidy code base than this. But we're explicitly talking about a project to just remove unused code here. Do we really need to get those many eyes all focused on this?
Nice blogspam, but why don't we link to the original blog post from Michael Meeks (Libreoffice developer)?
That's nothing. Last summer we has a wall taken down to extend the living room and we found Jimmy Hoffa inside it...
----------------------------------- My Other Sig Is Hilarious -----------------------------------
The design aesthetics are probably different because they intentionally make it different.
For most car brands, you can take the earliest cars and still see some of the design language reflected in the current models. In particular the front grills are usually easiest to recognize. This is done intentionally by the designers, even though nowadays car designers have no real connection to the brands; modern Lambourghini's are designed by a Dutch guy and Pininfarina (design studio for most of the Ferrari's) designs cars for other brands as well (i.e. Volvo, Peugeot, Ford).
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
I've seen chemical plants built with millions of dollars worth of unnecessary piping and valves, because the project timeframe meant that it was cheaper to install extra connections that might never be used and save engineering time than waste time re-engineering it.
If removing unnecessary items can save thirty thousand dollars (say) at the cost of three days, removing the cruft is only worth it if the delay costs less than ten thousand a day.
You're example is more along the lines of planning ahead to save money than leaving in stuff that is not used and is affecting performance. One company i worked fro, when they had a new office built, put in miles of cabling to met yet undefined connectivity needs because it is cheaper to buy the cable and install it before the walls go up than after.
Now, if they ran new wire and the old wire impacted its ability to send signals at high speed due to some inductive interaction, then it's like the dead code - something in place that has no use but is affecting current performance.
As you pointed out, in the end the decision is is the cost of removing the item worth the improvement we'll see?" With physical items, removal is also often driven by aesthetics; for example, once my deck supports were replaced I no longer wanted the temporary bracing left up simply because it look ugly, in this case the "cost" of looking tat an ugly deck was more than the real cost of removal of the temporary supports.
With coding, we often don't "see" the dead code - new features are added and code adapted to make it work; but there is no apparent "extra" code left behind or now rendered unnecessary by the changes. Plus, searching for it is boring compared to adding something else so coders move on to the next thing. Once performance starts lagging, then someone may look to "optimize" the code; because the cost of the performance hit exceeds the cost of optimization.
I'm a consultant - I convert gibberish into cash-flow.
"WORKSFORME" is a perfectly helpful response to "ITDONTWORK".
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
Actually, building materials (at least here in the USA) are dirt cheap.
Fucking well should be.
I've seen houses in the US built of plastic! Fuckers were all slowly falling apart as the UV degraded the PVC. The fake window wooden shutters were riveted to the fake wooden walls.
Watch this Heartland Institute video
... or maybe they belong to the previous owner's "missing" spouse?
The fact that software is easily and quickly replicated with perfect precision and little or no effort or time really makes it hard to make good analogies for it without resorting to Star Trek-style replicators.
Why not refer to biological systems? I wouldn't be too surprised if I have the code necessary for growing a tail hidden somewhere in my DNA, but last I checked it obviously wasn't being used. From what I understand, evolutionary processes as left to nature tend to leave a lot of cruft behind.
The problem friend is only 2K10 has support for ODT IIRC and you have NO way of knowing which version of MS Office the person you are sending that ODT to will have. Just in my family I have Office 2K on my netbook, 2K3 on my desktop and my oldest has 2K7 on his desktop and a copy of 2K10 his school gave him on his laptop. I have known businesses that have hung onto machines for lower positions like secretaries until literally the thing is about to die and often the software will be just as creaky, and with schools you have NO clue what the teacher will have.
So while your theory would be sound if you could easily know what the other guy is running you don't and THAT is a serious problem when you have business or grades depending on it.
ACs don't waste your time replying, your posts are never seen by me.
Excavating, as in a archeaological.
Under the false floors of data rooms across the world lie millions of miles of unused cable. Nobody is going to go in there and try to untangle them and risk disconnecting live cables.
I don't think this is the answer you're looking for, since you don't like Porsche styling and they too are from Stuttgart.
Space game using normal deck of cards: http://BattleCards.org
In the unlikely event that the code being refactored is poorly documented, (ahem, ahem), a programmer who implements faster, more compact, shinier code to replace existing libraries may choose to leave the old libraries in place, (still poorly documented), to avoid breakage and meet deadline. It's more like cities being built upon the remains of their older selves, rather than having a faulty construction manifest.
Actually an even better analogy was to use the car analogy...
Something matters, because there's (IMO) an enormous difference between the styling of Mercedes cars and BMW cars; it could just be company culture, but it could also be regional differences. After all, not that far away is France, home to several automakers, and I don't think anyone will deny that French cars have always had a very different design aesthetic from any German makes. Furthermore, there are cultural and linguistic differences between different parts of Germany. I'm just wondering if that's a factor in the differences between the various German automakers.
Interesting, I've always thought the Mercedes to be one of the ugliest.
But then I prefer pickups.
The only people bashing CISC are sheltered academics who have never been out in the real world or have heard of benchmarks. On my Core i7, for example, the benchmarks show that string instructions usually are the fastest way to do things.
A basic rep stos yields 25 GBps. So does the fastest recommended SSE method using movntps and prefetch. A RISC-style fill loop only manages 12 GBps. You can improve its performance by unrolling the loop four times, getting the same 25 GBps as a string instruction with a lot more code.
With a copy, rep movsl and movsq gives 3.2 GBps. movsb gives 2.5. A RISC-style copy loop gives 3.2 GBps. The fastest SSE copy loop with movntps and prefetch gives 6.4 GBps, at the expense of code size.
As you can see, for most applications it is a very good idea to use the string instructions. An SSE optimized copy may be useful if you're copying large blocks somewhere; otherwise don't bother. Outside tight loops, reducing code size will improve performance much more than directly trying to make it fast.
Bandwidth and memory are both pretty cheap
I don't know what kind of computer without a memory hierarchy do you have, but my instruction caches haven't been exactly cheap.
Ezekiel 23:20
In any project, 90% of code can be removed with no loss of functionality.
What material in your walls could be more insulting than newspaper?
An immured scientologist proselytizer that once came to your house to annoy you, only to never leave your house alive again?
Ezekiel 23:20
Can Office 2k open files saved in Office 2k10's default format? Having an older version of Word than the person sending you the file has always been a serious problem. And you can't tell even from the extension what format it really is. At least .odt is an actual standard.
Just a question. Isn't the extra coding necessary to get around patented code? And by removing the workarounds doesn't that leave the program up for patent lawsuits?
Jack of all trades,master of none
Under the false floors of data rooms across the world lie millions of miles of unused cable.
And many mummified mice, in my experience...
Having had to design changes to existing building systems that had lots of unused piping, ductwork, etc., I would say it's well worth the cost to get rid of obsolete work, rather than dealing with confusion figuring out what is active and what is necessary, or trying to snake new services through the existing congestion unnecessarily. It is relatively inexpensive to remove accessible items at the time they are made obsolete; the problems occur later when no one remembers what a particular item is for or whether it is still active.
I doubt very much that the Apache version of OO.org is going to do much improving at this point. Virtually all of the developers have already jumped ship for Libreoffice and I have yet to find anything that Libreoffice can't do that OO.org can. The reality is that they sucked in most of the patches from Go-oo and at this stage it's a superset of what OO.org was able to do.
You seem to be underestimating the number of patches and the amount of work that wasn't allowed into OO.org when Sun was running things.
Actually the answer is YES, yes it can, with the free converter pack. And I can testify that it works because my oldest always sends me DocX files from his Office 2K10 (He ALWAYS forgets to send as doc) and they open and display just fine. Now granted he's only a junior so his DocX files are just basic header and footer, no embedded tables or anything yet so I can't tell you how it will do with something fancy, but I CAN tell you that it opened a 15Mb business proposal I did with 4 other guys that had headers, footer, embedded tables, and frankly that amazed the hell out of me because I had 2K, one had 2K3, one had 2K7 and one had (2K6?) whatever the number for Office for Mac was around 2007.
So I can say that honestly I've never had a problem with Office since i started using it regularly with office 2K. Now maybe i just haven't tripped over the edge cases, who knows, but sharing both doc and DocX in 2K has not been a problem with the Office converter pack.
ACs don't waste your time replying, your posts are never seen by me.
And an excellent shoe company.
Lighten up Francis.
Let him who hath no extra wire hanging from the back of the rack cast the first flame.
Having to work for a living is the root of all evil.
Am I the only poster who will stand up for the response "Who cares"? The transistors analogy and, especially, the bridge-and-girders analogy, are just terrible analogies. To the customer, the cost of having legacy dead code sitting there on the hard disk is zero. The incremental cost of duplicating that dead code is also effectively zero. Software is unlike every other engineering problem for this reason. (And probably other reasons.) In other disciplines, it costs real resources and affects the working systems' performance if half of the material in the system overall is not used. Even if it were a problem, it's mitigated by, as mentioned above, the compiler not even compiling the dead code, in many cases. The actual cost is to those of us who have to maintain the code, or create new projects based on the code. As mentioned in other responses above, the learning time and workaround time is the bad part. But, we're basically paid by the hour, when it comes down to it, so wrap the code up and treat it as a black box and move on! Ship quickly! I know we all want elegance but our job, really, is to ship working systems quickly.
I built a swimming pool at my new house. The local code required the installation of a drain that runs from the sewer hookup in front of the house to the general area of the pool pump and filter -- where it connects to absolutely nothing.
Every attempt to figure out why this was required was met with a blank start and a vague mumble of, "It's the code."
"They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety."
(1) software has a potentially longer lifetime than hardware
Are you serious? There are plenty of bridges still in use from long before computing was even imagined. Consider these as a starting point. Respect to those engineers.
Ah, yes, I remember Adabas/Natural. The dmbs can't isolate updates from other processes before a commit. A mask to test alhanumeric input for a date format accepts some dates that raise an exception when the same mask is used to convert that input to a date field, the lack of stack traces punishes those who write modular code because you don't have a clue how the code where an exception occured was reached when calls get nested more than a few levels deep, exception handling is so limited that you can only crash or restart the application, no option to actually handle it and continue.
It's possible that it is good stuff, I may have misunderstood the German sense of humor.
Your emotions need better insulation . . .
hawk
That's nothing, when I renovated my basement I found the body of an old man, perfectly preserved. This was really upsetting. It was not the old man who vexed me, but his Evil Eye.
So-called "dead" code is usually the result of some feature that has been disabled or re-implemented. But keeping the code around as an example of how not to do things can prove very helpful to future maintainers. And if the code in question is ever needed again, you don't need to rewrite it from scratch.
In my case, there are several pieces of "dead code" features I tried and abandoned in the attributes of a business application model for MSS Code Factory. But although I left them in the model and they take up memory at runtime, they're concepts that I do want to follow up on again and tackle from a different angle than the former (failed) approaches to using that information.
There are also probably still a few unused methods and classes in the core code. I went through a huge cleanup effort over the past 18 months of a 15 year project, and got some incredible performance improvements along the way. The system now runs in under 10% of the time it took 18 months ago. Not a bad payoff, even though raw performance of the system was never a key requirement.
I do not fail; I succeed at finding out what does not work.
That would be relevant if the unused code was very tightly mixed with the used code.
In this case, if you look at the way they detect unused code, you'll see that all the code that was removed consists of whole methods and sometimes whole classes. This means that the vast majority of the code that was removed would probably never be loaded in the instruction cache anyway -- very small methods tend to be inlined by the compiler and so don't cause surrounding (and potentially unused) code to be loaded to the cache.
That's a pretty good idea; the big difference between software and biological systems, however, is time. For any macro-sized organism, it takes a long time for a clone to grow, unlike what they portray in the movies. It's perfectly possible to clone people, for example, but if you're 40 years old, it'll take 40 years to make a clone that looks like you (and by then, you won't look like that any more). Not only that, but biological systems aren't usually perfect in their DNA replication processes; errors happen all the time, causing mutations.
Software, OTOH, takes almost no time to replicate, only the time needed for the hard drive to read and write the pattern of bits associated with that software, or the time needed to transmit that pattern over a network, which is anywhere from milliseconds to a minute maybe, and errors are extremely rare and usually corrected by error-correction processes. This is very similar to the timeframe needed for Star Trek-style replicators; give it a command ("tea. Earl Grey. Hot."), and a few seconds later you have a perfect copy of that item in your hands. For a cup of tea, such ability is pure fantasy at this point, but for a piece of software that took many man-years to write, it's reality.
Not a good idea in a modern house, with wooden walls and drywall; he'll cause mold problems. It's better to take him down to your brick-walled wine cellar, where you can shackle him to a small alcove in the wall and brick him up. Even better if his name is Fortunato.
I did say that it would give them a "mildly" amusing story.
This is true in Europe too. I'm in Scotland. I live in a small but comfortable house I built for myself on my own land last year. It cost in total more than I earn in one month, but less than I earn in two. I'm a senior geek, I get a good wage - but I'm not a banker or a pop-star.
What makes housing expensive is partly labour, but it is mainly that planning policy creates a form of artificial scarcity, and partly also that government support for home ownership creates a speculative bubble, further inflating the cost. Housing is not intrinsically expensive.
I'm old enough to remember when discussions on Slashdot were well informed.
Frankly I don't know what's inside my walls, and I'm not sure I want to know.
Asbestos, lead, dead mice, and razor blades. (Old bath room cabinets used have a little slot to dispose of razor blades.)
The summary is clearly written from a user's perspective, and overlooks the reasons for deprecating code and features in a piece of software. There is a tradeoff when you make changes to a piece of software. If you are adding a new feature that does what an old feature did, do you leave the old feature? If you don't, it may break compatibility and interoperability with other software. If you do, you end up with deprecated code.
It's when so much time has passed that the reason for leaving the old code in place has dissappeared and the code and the reason for its existance have been forgotten that you run into trouble. This is where a code cleanup is needed. Deprecated code does not in-and-of-itself need to be removed, changed, or updated. But it should not be forgotten.
That's nothing. Whoever built the wall that exists between my office and the dining room left inside a leather dog collar and a half-dozen pork (I think) rib bones.
We've also found a cast iron floor lamp inside of a wall, as well as several hundred copies of the Saturday Evening Post that are positively impossible to drill through.
Pez dispensers seem so...basic.
Those aren't pork bones!
Dog collar.. animal bones.. hmmm.
Function level linkers have been around for a long time now. None of that dead code will end up in the binary.
Yes, I know most Linux distros ship LibreOffice now, but has anyone running Linux ever actually tried to install LibreOffice recently via the RPM downloads from the libreoffice.org? It's excrutiatingly packaged with so many flaws that it's embarrassing. Here's the list of what I have to put up with (at work, we use CentOS 6, which doesn't include LibreOffice):
1. There are *three* separate .tar.gz downloads (one for the core package, one for the main language pack and one for the help language pack (hmm...shouldn't the help language pack be merged into the main language pack)? Each of these .tar.gz's unpack into separate top-level sub-directories, just to inflame the awkwardness (I manually move what I want into the core package's RPMS sub-directory, but it's crazy to have to mess around like that). It should also be noted that the core package is dubiously labelled "en_US" but I'll get into that later on.
2. The RPM set has a major.major number *in the package name* (i.e. not in the version field where it should be). The lame official excuse for this is that you can install two LibreOffice versions side-by-side. How many normal end-users would *ever* do that, especially for two minor releases? Virtually no-one and yet this stupidity means that I can't upgrade any previous version easily. I have to manually uninstall all the old RPMs and then freshly install the new ones - completely maddening! Note that the developer version is named "LODev", so developers already have a completely differently named set of ("lodev"-prefixed) RPMs that can co-exist with the production release RPMs without the latter needing the version number embedded in the package name.
3. There are *two* core parts of the RPM package names: "libreoffice" and the bizarre "libobasis", so bang goes any chance to specify all the RPMs with a single wildcard.
4. I unpack the core package .tar.gz for, say, 3.4.4 and what do I see? Apart from confusingly having "rc2" in the unpacked dir name (yes, I know that the last RC = final, but it's still disconcerting), there's a shocking huge number of RPMs - 57 in all! For no reason at all, there's 7 "core" RPMs (core01-core07) when clearly they could put all that code into a single RPM. There are plenty of other RPMs that you suspect could also go into that single core RPM too.
5. Even worse, there are actually language-specific RPMs in the core package that make a mockery of having separate language pack downloads. Examples including *8* en-US RPMs (which I have to delete because I'm a British user and the en-GB versions are in the en-GB lang pack) and, even more astonishing than this is the inclusion of American English, Spanish and French dictionary RPMs, which I also have to delete.
6. There's also a "desktop-integration" directory in the core package, from which I'm supposed to install one of the RPMs ("redhat-menus" in my case) to get any sort of desktop menu entries, but shouldn't conditional code for this be included in one of the core RPMs and not as manually selected separate RPMs? I move the redhat-menu RPM up a level and continue...
7. Why is there a "testtool" RPM in the core package of a production release? I delete it and don't install it - normal end-users will never need it.
8. I then unpack the language pack download (en-GB for me) and have to manual move 9 en-GB-specific RPMs over to the core package directory (having had to delete the en-US equivalents first as I said earlier). Ditto for the help language pack (only one en-GB RPM this time).
9. I then have to run:
rpm -e `rpm -qa | egrep '(libobasis|libreoffice)'` to remove the previous LibreOffice release manually.
10. Finally, I get to install the new release with 'rpm -ivh libobasis*.rpm libreoffice*.rpm
No wonder almost all Linux users just used their distro's release of LibreOffice - the manual install is like having electrodes attached to your nether regions. I have never seen RPM packaging abused in such a significant way and it's especially galling when this is one of the most important apps - particularly for businesses - on Linux.
It's not an Engineering Discipline, no matter how many universities label it and the industry entitles it. It's Computer Science for a reason. Engineering disciplines are founded in Laws of Science. There is a reason Bill Joy has wished Software Programming could become an Engineering Discipline [he cited Mechanical Engineering specifically]--he's tired of the amount of wasted energy that the Art brings. That same Art brings a lot of creativity, but with a lot of undisciplined results and crappy code.
Think nothing of it. Print it and put it in your walls, and then you'll be insulated from it. Until it rains, at which point you may find yourself irritated by it again. At which point, you'll have some insulting filler.
Since when is that the problem with dead code being around in the first place?
Brian Fundakowski Feldman
Urban decay is a great example.
Take Philadelphia. We have structures that are more than 100 yrs old decaying in our city center! No one is funded to remove them. They will only be taken down when the real estate upon which they sit is worth enough to future developers to justify the cost of removing them.
It's the same with software. No one wants to allocate budget/man hours to this stuff until the cost/benefit ratio makes it worthwhile.
My God can beat up your God. Just kidding...don't take offense. I know there's no God.
I doubt very much that the Apache version of OO.org is going to do much improving at this point. Virtually all of the developers have already jumped ship for Libreoffice
Huh? Are you just quoting something you overheard or have I been duped by a massive conspiracy? Maybe you should start with learning the history of OpenOffice.org before making dumb statements. Sun fucked OpenOffice.org (deliberately). Many developers, who in total had contributed little of the code that made up OpenOffice.org (most came from paid developers) - but who never the less made important contributions to OOo, left when Sun cut them out of input into decisions. Now they contribute a large part of the code to LO (don't overlook Google's contributions). These are not secrets, or a matter "of opinion" - it's not fantasy football - hint: logs. Ditto with the amount of new code in OOo. Perhaps you confuse numbers with quality of product and discount the amount of catch up that the LO team had to deal with (especially shedding the GOo cruft) - a simple check of svn and git commits would dispel that myth. But I'm make it easy for you. It's not rocket science
They're both good products, OOo has more paid full-time coders, LO has more volunteers.
I have yet to find anything that LibreOffice can't do that OO.org can.
Compatibility with MS Office varies with both (particularly with Excel) - but in general OOo does a better job, PDF editing is better in OOo as is the ability to move from Write to Calc or Base. LO still doesn't handle complex math well (that should improve soon). And what of OOXML filters?
You seem to be underestimating the number of patches and the amount of work that wasn't allowed into OO.org when Sun was running things.
You think? I suspect you'd find the OOo team has done a lot since then if you didn't have such a fanboi attitude. And you're definitely overlooking IBM's contributions from Lotus. LO and OOo use different release cycles.
There will be a new release of OOo very soon - they've been busy the last few months doing just what LO is doing (it's almost like they talk to each other).
You're example is more along the lines of planning ahead to save money than leaving in stuff that is not used and is affecting performance.
Unused code doesn't affect runtime performance because, well, it's unused. At worst it consumes a little more memory unnecessarily which has the potential to cause more swapping (on Windows more than any other OS because it swaps "early").
It does cost you in development effort, though, as there's more code to look through that may be obscuring bugs in code that's actually used.
2K7 can also defiantly open odt documents and I think 2K3 can as well, I doubt that anyone is using an older version of office then 2k3 for business purposes.
null
The Historical Documents have the true story of a guy in England who managed to dispose of a number of bodies in acid, then poured the sludge into the drains.
Worked well until the police found a partial hip bone that blocked the sewer.
Don't be apathetic. Procrastinate!
Frankly I don't know what's inside my walls, and I'm not sure I want to know.
I have found out what is in mine and I will agree that you do not want to know what is in yours.
Libreoffice on FreeBSD compiles more reliably without the java dependency..
Java is used primarily for the database connector, so you lose the database functionality.
While the java dependency removal is on the road map. What to do about the advanced mail merge and wiki extensions that also depend on java is not yet clear (probably in the open source tradition, just drop support after a warning for one version about java support being deprecated.)
Work bio at MMWD
I don't get the analogy of the summary. It's like saying the deprecated code was never used in the first place. A better analogy is abandoned buildings, as they are the deprecated buildings. There are plenty of those around.
they've removed "over half the unused code".
but what the proportion of "unused code" was?
TFA says that the Libre Office had 5200 unused methods - is the nubmer high? compared to what?
thanks for explanation...
petr
umm, those aren't pork rib bones...
What?
There are lots and lots of documents that I can not see right in OO.
Try, for example a CV in European CV format.
Something that has very complex tables layout, like the a euro CV doc will fail straight.
Part of this is is due to 'reuse' and is a consequence of abstraction.
I've done remodeling. Electrical is always interesting. Electricians have no real organized system of working. You can't predict where a wire will run. Open up a wall, and spend a few hours going back and forth. (Does the line go to the light, then the switch, or is the switch 'beyond' the light. Can I splice in to *THIS* box and get power all the time, or will in be dead at one setting of the 3 way switches at either end.
If I reuse a 2x4 do I care if there are wiring holes I don't use? No. Do I care if there are nails clipped off and hammered flat. Not usually. If I move the sink, do I take apart the wall to remove the old sewer line? No.
Houses are not designed to be maintained. Nor are cities. (Try upgrading a city sewer system.)
Third Career: Tree Farmer Second Career: Computer Geek First Career: Teacher, Outdoor Instructor, Photographer.
And this is why programming is still an Art not Science. Eventually news methods of programming will have to be developed so it becomes obvious old code is really old (i.e. not useful). An analogy would be a house on a street that is boarded up and has unkempt lawn. The objects should keep track of last access and from whom so that judicial decision can be made to remove code: "Oops, the object you are requesting cannot be found. See object XYZ for more information." The software industry is still in the early days.
I don't know what you make in a month and the type of finish/comfort you have in your house, but typically I would say one would spend (in EUR) 5k in sand/cement/concrete blocks (foundation, support structures), 2-3k for steel/rebarb, 3-4k in wood (roof) + 2k in tiles, 8k in bricks, 3k in wiring/electrics, 9k for central heating, 3k for insulation, 6k for windows, ...
And these are just raw materials and very conservative estimates, no labor, carpentry (stairs/doors) or decoration.
That means you'd have to earn (net) more than 21k/month, but I would guess most (including myself) don't get anything near that amount.
Didn't work out so well for Toby Keith.....
http://www.youtube.com/watch?v=nOd2NuHgwew&ob=av2e
Another day, another update to a Google android app.
Our Genome also has a lot of unused codes which probably meant something at some point of time in evolution but, no more. It has genes that are deprecated. I think its very natural in an organically growing system.
If you have damp in your wall cavities then mold is not your only problem ....
Puteulanus fenestra mortis