1. Welcome to Smogon! Check out the Smogon Starters Hangout for everything you need to know about starting out in the community. Don't forget to introduce yourself in the Introduction and Hangout Thread, too!
  2. Welcome to Smogon Forums! Please take a minute to read the rules.

Issues concerning Project Concision

Discussion in 'Site Projects' started by bugmaniacbob, Nov 2, 2009.

  1. bugmaniacbob

    bugmaniacbob Floats like a Butterfree, stings like a Metapod
    is a Smogon Media Contributoris an Artist Alumnusis a CAP Contributor Alumnusis a Contributor Alumnus

    Joined:
    Sep 19, 2008
    Messages:
    2,305
    Firstly, I would like to apologise if this is not the time or the place to be bringing up doubts about decisions that have already made, or if you feel that this is not the sort of topic that I should be concerning myself with, given that I have never posted anything of note in the Inside Scoop before now. I cannot pretend that I have a fantastic track record of keeping to the new guidelines for analyses outlined in the newest batch of updates to be planned out. But I feel that Project Concision is a step in the wrong direction, as far as not only maintaining the quality and perfection that is remarkable to Smogon’s analyses, but also retaining the fullest support of the very best writers here on Smogon.

    If I may begin by quoting an example, we have here the Ninjask analysis that is currently on-site, written by myself, and underneath it the ‘concise’ version, written by Xia.

    I think that you will agree that, with no disrespect intended to Xia, there is a lot of information that has not slipped through the great net of concision. Relatively insignificant things, like a very rudimentary introduction to Baton Passing, for absolute beginners, descriptions of how each of the moves are used and benefit this Ninjask, plenty of examples of leads that Ninjask will have trouble with, and in particular, a good number of examples of Baton Pass recipients. Which adds up, if Xia’s calculations are accurate, to exactly 700 words cut straight out of the analysis. And in all honesty, if every single piece of information in the original analysis was put into the ‘concise’ analysis, I think that there would be less than a hundred words between them. (I haven’t actually tried this yet, but it’s only a rough estimate). And at that point, is there really any need to cut down on words? Cutting down on excess fluff is one thing, but cutting down on good information for the sake of aesthetics seems to be a ridiculous decision.

    Another point I would like to address is this, the seemingly universal agreement that short analyses are good analyses. Certainly, those that are straight and to the point gain merit over those that dawdle over certain points, or perhaps using three words more to address a certain point somehow makes an analysis boring or repetitive. But whenever I look at an analysis, what springs to mind is more often than not how much effort the writer has put into the analysis, rather than how pleasing it is to look at. Looking at Erodent’s Shedinja analysis for the first time, I profess myself quite amazed at how he managed it. Yes, Shedinja may need a lot of team options to be written about it. But Erodent obviously put the effort in to include as much of it as he could, leaving no stone unturned. And Theorymon’s essays never fail to astound me. Yes, they are long, but I cannot help feeling that the extra reading is worth it for those that want to read it. Smogon is publishing its analyses freely, but it doesn’t shove the analysis down their throats – they can read it if they want to, and the extra information should be there for those that want to read it – those people who are genuinely interested in the finer points of the Pokemon in question, not just the moveset. The contributors who write the analysis should not have to accommodate for those too damn lazy to read the whole thing – they should not be lazy themselves. This was one of my main thoughts while I was updating the Armaldo and Zangoose analyses – both were quite competitively viable Pokemon, yet neither seemed to have been written with enthusiasm or much effort (the latter in particular was limited on almost all of its sets to ‘Skarmory walls it, use Garchomp instead’). This seeming lack of interest (not an actual lack, but seemingly) is evident in some of the sections of the fully rewritten analyses, in particular the opinion section on Ninjask:
    The third and final point concerning these ‘concise’ analyses is the questionable message it is sending to those interested in contributing to Smogon. It is one thing to have a sizeable amount of quality control over the analyses submitted, in the interests of maintaining a reputation for eloquence and information in all analyses, and quite another to hold potential contributors on a leash and a strict word limit. Personally, I think that this unnecessary fluff may or may not be aesthetically pleasing, but they are very defining – those tiny little touches added, those 100 words in an analysis are the writer’s own special additions, that make it more an analysis and less of a bullet-point list. Perhaps they can be touched down a little, and in the case of the very best of concisions, such as Legacy Raider’s Zapdos, where a lot of effort is put into it, they may be the better for it. But everything in the world has a positive and a negative side to it – when you eat bread another goes hungry, by taking drugs for leukaemia you live, but lose your hair. These are considered acceptable because the positives outweigh the negatives, but we must ask ourselves whether this is the case with concision. The negative side being that, when crudely done, the analysis looks quite inexpert and is awkward to read, and even when expertly done, the results can look only so much better than what was there at the beginning. Simply put, where we could be allowing and advocating creative license to write well and informed pieces of work, we are restraining the amount we can put into our analyses and tying up resources and the very best writers trying to ‘write down’ work that took a lot of effort to construct, polish and finish. You may say that this fluff is unnecessary, and all we need for the analyses is not boatloads of information, but only what is strictly necessary. This would be perfectly acceptable, but… when you buy a car, do you buy one that is tarnished, scratched but essentially word-perfect, or one that has thousands of mod cons, polished surfaces, carbon fibre roof, every last nut and bolt given aerodynamic perfection? Which do you admire most? In the real world, this would all depend on the money you have, and your position in life, but then again, Smogon is rich. We have hundreds of talented writers, all capable of churning out accurate, superior analyses. Our position is as the premier competitive Pokemon site. We can afford the bells and whistles.

    If I may, I would like to add some suggestions as to how concision may be changed to a more suitable endeavour.

    • Firstly, if nothing else is done, concision should first be properly defined and regulated. The instructions are quite clear – the following analyses need shortening, so please feel free to do so, and suggest any others that you want to – yet everyone who starts on an edit seems to have a different idea of what this requires. Personally I feel that the best policy to pursue is Legacy Raider’s ‘chop and change’ – that is to say, editing the original analysis to preserve content and weeding out the little, unnecessary bits, rather than completely rewriting them, although as I have already stated, I still think that there are some issues with this method, not least that the analysis looks not altogether different from simply adding the process to grammar checking.
    • Setting an arbitrary word limit on analyses – perhaps 1000 words. This is by no means to long to read, and gives enough leeway for adding in any extra details that one may think necessary. The word limit would, of course, have to be increased for those with a presence in other tiers, such as Blissey and Porygon2, those with many different options and team structures, such as Heatran, or simply top-tier Pokemon that have a lot to be said about them.
    • Make it an official topic of grammar revision to cut out awkward sentences, and in general do what Project concision does now, only without a new subforum, without a new individual topic for each Pokemon, and with all the changes under the control of the original writer.
    • If it really, really concerns you, why not simply use hide tags? They cut out the aesthetic impurities, and anybody who wants to learn something can. Anybody who doesn’t can take a set and leave, without being blinded by science.
    Thank you for taking the time to read this.
  2. twash

    twash
    is a Site Staff Alumnusis a Team Rater Alumnusis a Forum Moderator Alumnusis a Contributor Alumnusis a Battle Server Moderator Alumnus

    Joined:
    Dec 26, 2007
    Messages:
    2,330
    I wouldn't argue that short analyses are "better", I would argue that they are more attractive aesthetically and more pleasing to newer viewers of Smogon, which is a considerable selling point.

    The general consensus in the project: concise thread was that the analyses could be shortened without losing information. If this isn't happening then either the aim of the project is wrong, or the project is not running as successfully as possible.

    Again, I go back to the keeping information idea. The plan was not to lose valuable information - just cut out the unneeded extra parts that don't actually help the analyses (besides proving the fact that the writer can well, write a lot).

    I'm just going to quote a small section of the above passage again:
    We face the same issue when writing longer analyses though - they can still be done awkwardly and inexpertly. Those analyses generally get re-done or helped with. Why can't we do the same with this? The simple fact is we have a quality check via proofreading and uploading to the SCMS - I don't see why that cannot happen with this project? On top of this, if we can make the analyses better, we should.

    An arbitrary word limit is awkward, because new sets are always being discovered, and some Pokemon will always have more sets than others. I really don't see the need for a word limit, we just need to make sure we are doing the best possible job.

    Personally, I don't like the hide tag idea, just because I don't see a reason to hide anything. The problem in my view isn't so much the aesthetical purposes of scrolling through to the next set, it's the aesthetical purposes of somebody wanting to actually read it. No player is going to want to look through 22 pages of text about Heatran just to find out information which could be found in less than a quarter of that. Why should we extend the time they have to spend reading for no real reason? Nowadays a lot of people don't have time to read 22 pages anyway; I think we need to remember that we are trying to help people, and I think chucking a ridiculous wall of text in front of them is clearly the wrong way to go about it. Why portray the same information in a ridiculously longer piece of text? Is it supposed to make the writer look clever or something?

    I would also like to point out that nobody wants to rewrite an analysis of 22 words completely, especially as it may lead to comments of how it isn't as in detail as the old analysis (when actually it is, just in less words). I think toning it down and shortening it all now will lead to less problems in the future.

    Overall, I think there just needs to be more people professionally checking that the same information is there. Comparison between the original and the latter should be paramount in my opinion; that way we can keep the same information but use less words. Thanks for posting this, by the way.
  3. diinbong

    diinbong *it's in you to give*
    is a Forum Moderator Alumnusis a Contributor Alumnus

    Joined:
    Oct 24, 2008
    Messages:
    956
    This is precisely it. I think the problem with this project may have been the "Project" tag it was given, which makes people assume as many analysises as possible need to be revised. In reality, though, only select few needed editing. This might have worked out better as a more private project to ensure that every write-up was as accurate and high-quality as its original (this project makes drawing concrete lines of what's acceptable and what isn't tough). If you feel information is missing in an analysis, though, just post in the thread saying so and giving suggestions. That's literally half the point of Contributions & Corrections.
  4. Colonel M

    Colonel M JESUS BIRD HAS RETURNED
    is a Forum Moderator Alumnusis a Contributor Alumnus

    Joined:
    Aug 5, 2007
    Messages:
    4,418
    This is exactly what I have been following, for the most part. Granted, there was a couple that slid through my mind (and I corrected the vast majority of them), but if I felt that they needed something shortened, I at least requested it. In fact, I have an ACTUAL RESERVE LIST (which has the analyses that need concising IMO) in the OP to prevent such a thing.

    The goal is to keep everything a bit more appealing to the eye. Yes, I don't mind reading something that has detail, who doesn't? It's stuff such as Scizor (and guess what, I wrote about 30% of that analysis, btw so I know I'm victim of it), Heatran, and others that really need the shortening. Of course, if you feel that there is actual information MISSING or NECESSARY, then simply point it out in the thread.

    I'm trying my best to look through these, honestly I am. It's not as easy as it sounds. Take SD Rayquaza for example. Though it seems very long, try making it shorter without making it sound extremely dull or without information. Which is why it's not just my goal, it is other's to help point things out. I apologize for the further inconvenience otherwise.
  5. mtr

    mtr

    Joined:
    Jan 31, 2009
    Messages:
    1,773
    While concerns about losing information are valid, there are cases which, to be very blunt, read like the work of a college student trying to get past a word limit. Ergo, they could be summed up much more succintly. For example, take a look at Curselax below.

    I can honestly state that I wouldn't go to the effort of wading through all these unnecessarily long sentences just to get information that could be given to me in a few sentences on IRC or (god forbid) the shoddybattle chat.

    This is my concise version of it.

    I understand that I am sometimes also guilty of having too-long analyses, such as Tentacruel, but the point still stands. Sometimes, we get too prosaic with our writing, and it needs to be trimmed down. The hope is that the trimmer can see what needs to be kept (calculations, discussions of other threats, etc.) and what needs to be chucked (which usually has its basis in sentence structure).
  6. bugmaniacbob

    bugmaniacbob Floats like a Butterfree, stings like a Metapod
    is a Smogon Media Contributoris an Artist Alumnusis a CAP Contributor Alumnusis a Contributor Alumnus

    Joined:
    Sep 19, 2008
    Messages:
    2,305
    Firstly, I would like to say that I was nearly finished making a long reply, when naturally my computer closed down the internet. So, in an act of supreme irony, I will have to make this post concise.

    That's true, but having a short analysis is not the only option to have selling points. There are also more advanced options to consider, for those that are slightly further on than being a newer viewer and also willing to read the analysis, vague descriptions of the set and tactics commonly used are simply not enough. Perhaps it would be better to split basic tactics and advanced tactics into two separate paragraphs or tabs, I don't know... in extreme cases, such as perhaps the Heatran analysis, I would tend to agree with you that they need trimming. But I would only have this reaction for those that took up significantly more than a whole browsing window on my screen (such as the Heatran analysis, but not incorporating Zapdos, for example), which many of those submitted for concision do not. Otherwise, we are losing good information for the sake of aesthetics, which I do not like the sound of.

    I am not going to refer to the overall drive of the project, as I do believe that the job it is doing is well executed and efficient in most areas, nor the philosophy of the project itself. I shall only say that the analyses could be shortened without losing information. Whether they could be summarised without losing information is another matter entirely, and I shall not pretend that I would consider a summary that incorporates all the information from the original to be a complete logical paradox.

    There is, as of now, no guidelines or help beyond the personal opinion of those involved in the project as to what these 'unneeded extra parts' stretch to in terms of legitimacy. And thus, almost all of the concisions follow the editor's own opinions on it - some are completely summarised, some trimmed down to an acceptable length. If there was a universal standard to this I would not object, but there isn't, or does not seem to be. In any case, as I have already mentioned, summaries have a remarkable tendency to lose good information, whereas trimmings, while keeping all the information, do not seem to satisfy the requirements for the concision of analyses in terms of their length, not making much difference especially to the much longer ones (here I must draw your attention to the Blissey concision).

    Fair enough, but the summarised analyses lack the detail and information of the finished analysis. A much bigger issue I would have with this would be the difficulty by which the proofreading system is woven for concise updates. It's all very well to suggest the addition or removal of something or correct some grammar, but the quality check for concise analyses is quite different - 'suggest a sentence to be picked over from the original'. These may seem quite similar, but I would consider them worlds apart - one is a top-down approach, the other a bottom-up approach. I also would not agree that writing a similar, generalised analysis is necessarily better than a more detailed but wordy one (and neither do you, if your first sentence is to be believed), especially when the wordy one can be cut down by trimming rather than summarising (I tend to favour the former as the most practical, if not the most aesthetically pleasing, option).

    Toning it down I have no objection to, but summarising is what I have a real issue with. The problem being that, from what I have seen, simply toning it down is not enough to make it short enough on its own, especially when the original analysis is, as you say, 22 pages long. At that point it seems logical to summarise parts, but the problem comes, not only in the form of consistency, but also in deciding where and when an analysis requires summarisation to make it acceptable.

    I completely agree. But I would stress that the focus should be primarily on those doing the concision, rather than those professionally checking it - keeping the same information is important, but the job of doing so is made easier for those trying to correct it if they have the information already in front of them, that is to say, if the analysis has been concised rather than summarised.

    I don't honestly believe that suggestions are enough to maintain the entirety or majority of the information within the analysis, especially when contributors take the word 'concise' literally to heart. In my Claydol Update, there are several requests for it to have the fluff removed, but few actual recommendations for how to go about this. I am not complaining, as in this situation it is completely my responsibility to get the analysis in perfect condition, but I feel that this effect can be felt at the other end of the scale as well, particularly with regard to the suggestion of things to keep in. Try as you might, you cannot quite readdress the specifics of what you are trying to say while keeping it concise, or so I have found, and for me they are some of the most important parts of an analysis

    I do not doubt for a single second that certain analyses require shortening of some description, nor that you are in any way less than committed to the project, nor that the guiding aims of the project are fundamentally flawed. The issues I have with the process of concising analyses at present, put simply, are Summarisation, Information and Continuity. Summarisation and Information are pretty much one and the same, that rewriting analyses from scratch and then attempting to splice in extra bits that are needed is a bad way of going about making longer analyses more focused. Not only that, but also incorporating the issues that arise from shortening analyses without direction, such as how short exactly, and debate over what or what not to keep. Continuity issues refer to the slightly mishmash way that the community as a whole approaches Project Concise, without any real starting guide beyond 'Pick, Shorten, Post'.

    Some of the stuff you have missed out on:

    • Curselax history
    • Why you should use Curse on Snorlax rather than anything else
    • Double-Edge comments
    • More emphasis on the context in which different options are advantageous
    • Curse does not boost Special Defence – your concision of this sentence means that the meaning becomes less clear
    • More specific reasons for the odd HP
    • How, precisely, Gliscor/Heracross act as notably good partners for Curselax
    Plus all of the little details that you may have considered irrelevant. The analysis itself is very good, but in comparison with the original Curselax analysis, yours seems generalised and non-specific – the concise version flows but does not strike, so to speak. This seems to be the problem with all ‘summarised’ analyses – they can and will lose information in the transition, and even if you included every last bit of information in the analysis, I still firmly believe that you will have accomplished very little that could not have been done by simply trimming down the analysis – that you will have lost little more than a hundred words for all your work – quite disproportionate to the amount of effort you put into it.
  7. Colonel M

    Colonel M JESUS BIRD HAS RETURNED
    is a Forum Moderator Alumnusis a Contributor Alumnus

    Joined:
    Aug 5, 2007
    Messages:
    4,418
    I'm not wasting much more time with this, but I might as well nitpick here.
    This isn't necessary.
    Also question how "necessary" this is.
    Name someone who's honestly used Double-Edge? This can be covered in Other Options.
    - Traditionally, Rest was used in the third slot, but in the current metagame, one cannot afford to give free setup opportunities to enemy sweepers, so SelfDestruct is an excellent option to take out an enemy Pokemon with you. For the last move, Earthquake can be used against Tyranitar and Metagross, but Crunch, despite being useless against the two, is necessary to avoid being walled by Gengar and Rotom-A (although Gengar without Focus Blast or Perish Song can be PP Stalled by Rest). Fire Punch can be used for Ghosts and Metagross, though Tyranitar still switches in easily. -

    What more do you want?
    - Curse boosts Snorlax's Attack and Defense, making it better at taking both physical and special attacks. -

    ???
    - The EVs are designed for optimal defensive efficiency, with the odd-numbered HP to ensure that Snorlax avoids being worn down as much by residual damage, which is dealth in eigths and sixteenths. -

    Seems pretty dead-on to me.
    Perhaps one of the few things I agree on, though he did put some emphasis on it:

    - Despite the use of Curse, physical Fighting-types, such as Lucario, Machamp, and Heracross, can still defeat you with Close Combat or DynamicPunch. Therefore, the use of Pokemon that resist Fighting-type moves can can handle such threats, such as Dusknoir, Rotom-A, Gliscor, or Weezing, is advised. -

    I did at least try to emphasize that the Team Options stuff didn't need a whole lot of cutting unless the writer felt it was absolutely necessary. Though I do agree that it should have a little bit more reasoning (i.e. Gliscor can lay Stealth Rock, Rotom-A can set up Reflect and Light Screen and possibly Will-O-Wisp, etc.).
  8. mtr

    mtr

    Joined:
    Jan 31, 2009
    Messages:
    1,773
    I should probably mention that I'm still undergoing edits on Curselax. However, let's zoom in on this particular paragraph.

    Why is it necessary to have all these extremely long and convoluted sentences? I stand by my previous point that many things that are too long are more a result of grammatical structure and redundancy and less a matter of content.

    Concise version.
  9. bugmaniacbob

    bugmaniacbob Floats like a Butterfree, stings like a Metapod
    is a Smogon Media Contributoris an Artist Alumnusis a CAP Contributor Alumnusis a Contributor Alumnus

    Joined:
    Sep 19, 2008
    Messages:
    2,305
    And that section was probably the best instance of concision that you had in the entire piece. Notably, because you seemed to have simply trimmed the longer sentences rather than completely rewriting them - I notice that though it is not all that shorter than the original, all the information is still there, which is great.

    It certainly isn't necessary to have long or convoluted sentences - I stand by my point that trying to explain this more thoroughly, adding the little bits of information that may not matter to some people, makes the analysis much better than keeping a limit of 300 words your first priority. Sure, some of these analyses may have to be shortened, but a lot of the ones I've seen could be kept clean by just trimming.

    In all honesty, by that reasoning you could claim that most of the analysis isn't necessary - if you cut out everything that seems obvious you aren't left with much. The Curselax set played a big part in previous generations, so why should it not be included? It's interesting, and it doesn't need more than a simple sentence or two to explain. The bit about Curse being used rather than anything else - If Tyranitar could use Curse to a greater extent, that would surely be mentioned. Snorlax has good Attack, great natural Special Defence, and doesn't mind the Speed loss. Which is why Curse is better on Snorlax than say, Aerodactyl.

    Fair enough.

    I was actually referring more to the differentiation between Body Slam and Return there, but in any case you could say that Selfdestruct is better if you require a reliable revenge to most set-up sweepers, or else don't consider Curselax your main sweeper (whereas Rest would be better if you did, most likely). You could also discuss the main team types that either move would be advantageous against. You would probably disagree on this point, but it doesn't really matter that much.

    As in, "making it better at taking both physical and special attacks", which makes it sound as though Curse raises Special Defence. I would have put, "making it better at taking physical attacks, which in addition to its already impressive Special Defence...etc."

    Perhaps it's just me, but "residual damage, which is dealt in eighths and sixteenths" sounds just a tad ambiguous - If you stated that Sandstorm took away an sixteenth, and Stealth Rock an eighth, etc, then it would start looking a lot less confusing to newer battlers.

    Anyway, most of those were just slight nitpicks, I didn't want to start a war over them. I suppose this slightly illustrates one of the points I have been trying to make - If we have unified guidelines to the extent at which information should be removed, and a consistent degree to which analyses are trimmed down, the project and the analyses would be much the better for it.

Users Viewing Thread (Users: 0, Guests: 0)