FYI: Creating circular references is a perfectly OK thing to do.
in [JavaScript]
|
From: David Mark on 29 Jul 2010 02:04 On Jul 29, 1:20 am, Alan Gutierrez <a...(a)blogometer.com> wrote: > David Mark wrote: > > On Jul 29, 12:51 am, Alan Gutierrez <a...(a)blogometer.com> wrote: > >> David Mark wrote: > >>> On Jul 28, 11:43 pm, Alan Gutierrez <a...(a)blogometer.com> wrote: > >>>> David Mark wrote: > >>>>> On Jul 28, 7:33 pm, Alan Gutierrez <a...(a)blogometer.com> wrote: > >>>>>> David Mark wrote: > >>>>>>> On Jul 28, 5:35 pm, Alan Gutierrez <a...(a)blogometer.com> wrote: > >>>>>>>> Richard Cornford wrote: > >>>>>> I am only interested in driving home the point that > >>>>>> you have yet to refute. > >>>>> You have no *relevant* point. Please drive home now. :) > >>>> I also say the point is relevant, and you have yet to refute that. > >>> Refute that your point is relevant? > >>>> You > >>>> can't refute the argument so you keep misdirecting the argument. > >>> I'm not arguing with you at all. I've long-since informed you that > >>> I'm uninterested in this irrelevant offshoot. > >> I have no evidence that the point is irrelevant. You're waving you > >> hands, but you are not being reasonable. It is now, in my experience, a > >> place you retreat to when you cannot offer reason, to simply declare the > >> point irrelevant, or say that someone else made the point for you. > > >> I can only assume that you can't make a point on your own. > > > Making assumptions about me is a slippery slope. > > >>>> You > >>>> actually have an interesting habit of removing the part that you're > >>>> supposed to refute, and then picking up on a word, "seems" or "iPad" and > >>>> setting up your straw men around that. > >>> No, I rarely snip anything, much to the chagrin of some regulars > >>> here. I did in this case as I had no interest in the your continuing > >>> (and seemingly unending) attempts to drive home a point that is > >>> irrelvant to the discussion. > >>>> So, I invite you to refute my point that platform is a choice, therefore > >>>> is is just as logical to choose the platform of the future instead of > >>>> the platform of the past, > >>> That's easy enough. You don't need to choose at all. You could have > >>> saved some time by reading the post I cited. ;) > >> Compromise is inherent in software. More platforms means more money. > > > So narrowing your choices means less money. ;) > > Yes. Exactly. Except by money, I mean costs. Should have made that > clear. I believe your starting to see my point. Not at all. I thought you were referring to the money you hope to make. > Is there hope that you > might see the reasoning, yet? Again, you should really read that post I cited. It describes how I can retrofit any existing site to work in narrow and animated native- like fashion in any of those devices, without the desktop user even realizing it. Of course, the exceptions are very bad sites that use tables for layout, huge bloated and/or browser sniffing scripts, Flash, etc. Those would require actual work. It doesn't cost any additional time if you do it right. Though if you don't know how to do it right (and can't be bothered to learn), I guess you are screwed (like those Flash-y, Ajax-y, magazine-layout- like sites that have been the rage for years). > > >> Conversations with you are a luxury I afford myself. > > > Well, if you are ever feeling the pinch, you might want to try out our > > utilitarian Jorge line. > > >> I see no profit in > >> them, time or money, other that perhaps to get you open your mind to new > >> ideas. > > What are you my therapist now? :) > > I'm having a go at reasoning with you. You said you wanted to open my mind. Thanks, but I think you have the wrong number. Please hang it up.
From: Scott Sauyet on 29 Jul 2010 08:02 Alan Gutierrez wrote: > Scott Sauyet wrote: >> [ ... ] Jorge was arguing much further that taking >> advantage of this and writing scripts that would obviously fail in IE >> would be an effective technique to either move people away from IE or >> convince MS to fix its browser. That argument is much less clear. >> Most of the people who've tried that, I imagine, have found themselves >> fairly disappointed in the results; unemployment is rarely pleasant. > > Moving away from an end of life platform and toward contemporary > platforms is a valid strategy. Moving off of vulnerable and unsupported > platforms toward contemporary platforms is a valid strategy. Dwindling > support for a platform expedites the drain. Certainly, Microsoft must be > feeling pressure to catch up to Safari and Chrome, which are backed by > serious competitors. I doubt Microsoft is overly worried about better browsers stealing its market share. They've never found a way to make the browser earn money for them. They're probably better off worrying about Google Docs and Linux and the dominance of the iPhone, issues which cut into their bottom line. The point has been made repeatedly in this thread that there are many corporate environments in which some version of IE is the only allowed browser. There are still a number, I believe on IE6! That simply means that an app that doesn't work in IE is simply not going to be used in these environments. That is changing, albeit slowly. Firefox has definitely made inroads in corporate environments, and I imagine Chrome will as well. I recently completed an assignment for ESPN modifying an in-house Firefox-only app! But the reason it was FF-only was that the initial developers weren't competent enough to make it work cross-browser. Fixing the issues that caused that were never part of my task, although they were not severe. So this does happen, even in large corporations (ESPN is owned by Disney.) But it's not a major factor as far as I can tell. And I doubt that Ballmer is worried by people like Jorge. >> If a developer wants to treat an improper garbage > collector as an issue they do not want accommodate, I can see how it > will reduce the cost and complexity of their offering, and their > platform will only grow as time goes on. Yes, but they have to be willing to forgo a large portion of potential users right off the bat. Even if they are targeting, say, the iPad, should they make the decision up front to make it more difficult to later port the application to a wider environment? Usually, I contend, the answer should be no. > Otherwise, they are going to have a legacy user base, and it is hard to > lose customers that you've begun to support. There will be plenty of > employment for people who skate to where the puck will be. You seem to be misunderstanding. The issue is not writing in a way that will only work in IE; it's writing in a way that will *also* support IE. Others in this thread have rightly argued that circular references with host objects is a bad idea regardless of IE's particular issues. But the main point is that there are techniques almost as easy to use that don't cause problems in what is still the most widely-used browser. Why not use them? -- Scott
From: Richard Cornford on 29 Jul 2010 09:27 On Jul 29, 12:41 am, Alan Gutierrez wrote: > Richard Cornford wrote: >> Alan Gutierrez wrote: >>> Richard Cornford wrote: >>>> Alan Gutierrez wrote: >> <snip> >>>>> Both of which arguments? Matt is saying that the other browsers >>>>> do not support ActiveX. Maybe I misunderstand why this is >>>>> relevant to Matt. I assume he is saying that is an argument >>>>> in favor of IE. > >>>> It isn't an argument about IE, it is just an observation about >>>> IE. There are business in the world that have intranets on >>>> which they have browser-based applications that they use in >>>> order to conduct their business. Some of these applications >>>> use ActiveX (because ActiveX can do things that ordinary web >>>> browsers just cannot, in some cases) and these businesses will >>>> not be giving these applications up because they need/want >>>> them. So in these environments the browser installed on the >>>> business's (likely 'locked down') desktops will be IE. > >>>> If you want to sell into that sort of environment then you >>>> have to cope with IE, because if you don't the sales will go >>>> to your competition, because the client dictates the >>>> environment. > >>>> That is the reality in web application development, but it has >>>> obvious implications for the general web, particularly >>>> e-commerce. If someone working for such a business is going >>>> to do a bit of online shopping during their breaks (and there >>>> is no point in pretending that they don't) then they will be >>>> using IE to do it. Now the online shop that doesn't support >>>> IE is losing the business to its competitors that do. And >>>> remember that these potential customers are, by definition, >>>> in employment, and very often in well-paid employment >>>> (exactly the sort of customers most business want). > >>> If it is the case that you are deploying a proprietary technology >>> like ActiveX, > >> I am not. What I am doing is designing/building web applications >> for business use. > >>> then it is the case that there are situations where the >>> application matters more than the browser, > >> Possibly, but those decisions are made by the client's IT >> department, and are not open for negotiation. > >>> so requirements dictate the browser instead of the browser >>> dictating the requirements. > >> For me a requirement is that the browser is not dictated, but >> rather accommodated. > >>> Therefore, if I'm building a web application and I want >>> to target the iPad, > >> You want to target? So this is purely personal project where you make >> the business decisions? > > On every project I make the business decisions. So you have no external clients? Because if you have external clients then the business decisions that relate to browser software design are their responsibility (as it is their business). > I make the business decision to take on a project. If the project > is burdened by arbitrary requirements, I'll pass on the project. That is a completely different sort of business decision from the decision to create something that 'targets the iPad'. Yours is in internal decision, and it is entirely your own business if you chose to walk away from proposed projects. But where, for example, a project is to create something that sells widgets to punters the impact on the success of the outcome at achieving its purpose is very much the business of owner (funder) of that project, not the developers working on it. If the design is going preclude some percentage of customers 'up front' that will impact on the owner's business, and the decision to do that is then the owner's to make (hopefully with some (informed) input/advice from the developers). > You seem to bring it back to your choice of platform Haven't I made the point that in the context in which I work the choice of platform is effectively made by client's IT department (or rather was made by that department at some point in the past). > and your business It isn't my business. I am an employee. > and your personal preferences. My personal preferences have very little impact on what I do beyond deciding the architectural questions and some of the coding style in the client-side code of the web applications that I work on. > The fact remains that Jeorge is correct in his argument that, > if you choose not to support browsers that do not provide a > proper garbage collector, then these issues of circular > references do not matter. Correct, but only trivially so. It would also be true that if you abandoned browser scripting entirely then these issues of circular references no longer matter. Obvious, but not worth saying. The relevant question is what consequences will follow if you set about exclusively creating browser software that will be problematic in IE. And the answer is that in today's market if you are 'targeting' iPads then it won't make any difference to anyone, but if you are selling (software to business or products to customers) or advertising/ promoting (yourself, or raising revenue through) then there will be at lest some negative consequence, ranging to the extreme where some projects/businesses will be rendered non-viable. <snip> >>> Let's say I want a single purpose front end and the entire >>> organization is committed to running one application and >>> wants the shortest path to a working application. So, I >>> choose Chrome and HTML 5 and a JavaScript implementation >>> that has property garbage collection. > >> Fine, you can say whatever you like, but if your customers >> will not play ball then they won't be your customers. That >> may not always matter but sometimes (indeed often) it will. > > Obviously. This also makes my point. If they are supporting > a platform that I do not support, then we both go back into > the market to find a better pairing. But having to go back to the market represents a failure to sell, and the handing over of potential business to others. > You conceede the point here. You admit that I am right, > that Jeorge is right. So the discussion can end here. You > can specialize in maintaining the environments of the past, > I can specialize in the creating the environments of the > future. Either would be crazy because the money is in the market of the present. Propaganda aside, the "environments of the future" is a risky game. I have been at this long enough to see a precession of future predictions. So far reality has not tended to deliver on any of them. Eight years ago there were plenty of people buying the idea that the future of the Internet would be IE only, and so we did not need to bother with any of this 'cross-browser' stuff. And there were even people proposing that if nobody bothered with this 'cross-browser' stuff then the future of web would be IE only (and everything would be easier). >>>> And if you were never in a position to say either? > >>> And if you were always in a position to say both? >> <snip> > >> "Always"? That sounds like a very specialised context that most >> people are unlikely to find themselves in (often, if ever). > >> Matt's point in response to Jorge's suggestion is an >> observation of the market as it is today (and some explanation >> of why the market is the way it is). People will act to service >> that market, that will undermine Jorge's grand scheme for >> dictating browsers, and since the scheme relies on everyone >> going along with it, it must then fail. > > Jeorge's grand scheme is to build applications that are not > burdened by the weight of a 10 year legacy. That rather assumes that there is 'weight' to do the burdening. It has been pointed out that it is not actually that difficult to cope with IE's circular references issues, including coping by adopting a practice of never creating circular references (and so never provoking any consequences of doing so). > It rules out a lot of shops, but not all of them, and if there > is a market large enough to pay his bills now, My employers would have to be pretty certain that there was such a market if they were going to gamble a multi-million dollar global business on it. To date, one client has enquired about using Chrome, one about Safari, none about Firefox (a little surprisingly) and all the rest (actual and potential) are IE only. > it will only get larger as time goes on. While the market for browser scripts that also accommodate IE will not get any smaller. So, you can go along with Jorge's grand scheme, and in 4 or 5 years, if things do go to plan, you will be back to competing on a level playing field. Richard.
From: Ry Nohryb on 29 Jul 2010 10:17 On Jul 29, 2:02 pm, Scott Sauyet <scott.sau...(a)gmail.com> wrote: > (...) And I doubt that Ballmer is worried by people > like Jorge. It's hard to tell what worries him the most when he's not on cocaine. Is it the shrinking Office market ? Their catastrophic failure in the mobile OS market ? Apple? The ~ null penetration in web servers ? Linux? The plummeting market share of IE ? The Chromes, Safaris, Firefoxes and Operas out there? The Zune's failure? The iTunes music and app store? That nobody uses Bing ? Google? Is it Microsoft's total incompetence what worries him the most? Or his own stupidity? Who knows... But yes, certainly he shouldn't be worried by the people who thinks like you, but I'm not so sure about the people who thinks like Jorge. > (...) Others in this thread have rightly argued that circular > references with host objects is a bad idea regardless of IE's > particular issues. (...) Is that a saying ? Why is it a bad idea? Because they (well, it's been only Asen) have said so but have given no proof nor reason for it to be so, except that IE leaks. Because it isn't bad per se. CRs aren't neither good nor bad. It's just that the word has spread that CRs leak memory, which is false. IEs are what are a bad idea, not the CRs. > But the main point is that there are techniques > almost as easy to use that don't cause problems in what is still the > most widely-used browser. Why not use them? Because there's no need. Because there are much more interesting things to code and think about than workarounds for Microsoft IEs' bugs. And because we should not circumscribe the web to the least common denominator (for cross browserness) when the least common denominator is a fraction of what it would be if it were not for IE. -- Jorge.
From: Ry Nohryb on 29 Jul 2010 10:40
On Jul 29, 12:25 am, "Richard Cornford" <Rich...(a)litotes.demon.co.uk> wrote: > > (...) and since the scheme relies > on everyone going along with it, it must then fail. The soldiers knew well -first hand- what was going on. Civilians knew it too because they were being told by their parents/brothers/sons/ husbands soldiers, but mostly, (66 millions of people!) thought/ pretended/believed there was not much that they could do about it. 1939-1945. History lesson: no matter what, only the right thing to do is the right thing to do. -- Jorge. |