I don't know enough details about the modern Linux kernel to comment usefully on that specific case, but certainly as an industry we have a lot more experience with hot reloading techniques today than we had a few years ago, and certainly a lot of devices don't have or need anything like the complexity of modern desktop/server operating systems.
I deliberately wrote a figurative 99% of the time rather than 100% before, because in the real world you might still have a situation where you really can't update the very heart of a system without taking it out of service at least briefly. Even with perfect software design, sometimes the hardware limitations might get you. But with reasonably designed, modular software architecture, we really shouldn't see devices becoming useless for extended periods anything like as often as we do in practice at the moment.
I don't have any special insider information or anything. I just looked up the most recent official data I could find for Adobe, and highlighted a few key points that stood out to me for comparison.
Since you asked, my guess is that there's a simple explanation for Office 365, something like the following.
Firstly, Office 365 is on sale at the same time as a more traditional version in Office 2016. It's reasonable to assume that some people and organisations prefer the subscription/rental model, while others prefer a more traditional permanent purchase.
Now, some of the pro-subscription group have already subscribed, so that leaves fewer new subscriptions from among that group now (and more renewals). Also, some of the pro-purchase group will presumably have updated to 2016 recently, so they may be less inclined to throw that investment out by signing up for 365 so soon. Both of these point to a drop in the number of new subscriptions.
To reverse that trend, Microsoft would need to find more people who want to have Office and/or get more existing customers interested in subscribing instead of purchasing outright. However, unlike say Adobe or Autodesk, Microsoft can't easily force their existing customers' hands by going subscription-only. It's a relatively mature and saturated market. Existing versions already do most of what most people need in an Office-style product, so few people switch or upgrade once they've got their copy. There's more innovation in related products nipping at the edges of the market or aiming at specific niches, but O365 isn't really addressing that part of the industry. And of course the elephant in the room is that there is credible, direct competition to Office, so if Microsoft tried to force happy, purchase-oriented customers to move, they might find Office 365 was not where those customers moved to.
So, as long as they want to sit on their 800lb gorilla cash cow without making a game-changing new release, MS might just have to accept that a lot of customers still value Office enough to buy new copies but many don't want to rent. If so then, as our showbiz friends might say, they need to give the people what they want. That inevitably leads to some slowing down of new subscriptions as the market reaches an equilibrium between the two models.
As I said, I have no special insider knowledge to back this up. I'm just applying Occam's Razor.
Here's an investor presentation from last week (PDF). It's a long document, but the following is mostly derived from a couple of slides at the bottom of page 3.
It looks like their total revenues for Creative Cloud dipped a few percent and then recovered again over the period 2012-2015, and as of 2016 their annual recurring revenue for that area is up to around $3.5B, compared to annual revenue of around $2.5B back in 2012 when their subscription model was starting up.
Over the same four-year window, it appears that their subscription ARR has been increasing roughly linearly, while their non-subscription revenues are fast approaching zero.
In short, it looks like they are now better off than they were four years ago in terms of annual Creative Cloud revenue, by about 40% if they maintain their current subscription level.
Another figure they mention is current year-on-year subscription growth of 46% outside the US. However, they are deafeningly quiet on what proportion of their overall market that represents or the equivalent figure for US customers. Their overall growth rate is clearly far less than that, so it could be that they're successfully expanding into foreign markets and that's helping to drive their overall subscription growth (probably a good thing for Adobe) but it could also be that sales in foreign markets are covering up a significant reduction in the US as increasing numbers of US customers are cancelling their subscriptions (probably a bad thing for Adobe).
It's also difficult to tell how many subscribers they actually have, since there doesn't seem to be any breakdown of which of the available subscription plans are generating how much revenue or what sort of effects they see from volume licensing, subscribers from different countries, or subscribers paying in different currencies. If we guess an average subscriber is worth about US$500 per year to them in revenues, that would give them around 7 million current subscribers, but this could obviously be way off if say most of the revenues are actually from enterprise customers paying far less than the headline per-seat prices with their volume deals.
Obviously the devil is in the details but NFC keycards aren't going anywhere (no changing locks and lost keys)
OK, I'm with you so far.
and internet aware locks are the obvious next step of convenience and cost cutting.
::boggle::
Even some of the cheaper hotel chains here in the UK now routinely have machines that let you check in without staff intervention, including coding your keycards for you. It takes a few moments. It is not at all obvious to me that Internet-enabling anything about this process would be either more convenient or cheaper for anyone.
Maybe. The article doesn't say, but if the attack truly did lock the doors from inside then clearly they have other problems, for sure. At that point, you're not talking about vulnerabilities to hostile parties any more, you're talking about basic reliability and safety concerns.
The article doesn't specify the exact system or how it was compromised, so unless you have some other source to share, none of us know whether the devices that were compromised in this specific case were directly Internet-connected. Some modern hotel systems are. It could also be that the repeated hacks in this case accessed the room key system indirectly via some other system that was compromised first. The fundamental issues raised are the same either way.
Who thought it was a good idea for essential systems like this to be online in the first place?!
This is why the Internet of Things is such a horrible concept. Most things don't need to be online and connected to everything else, and the cost of trying to be trendy is huge increases in risks to the privacy, security and reliability of everyday items.
Closed networks do just fine for these kinds of systems, don't actually need to cost that much more, and have none of the vulnerabilities.
But as a web developer, Google Chrome and the other "evergreen" browsers are a nightmare for stability. While having support for newer HTML or CSS or whatever is sometimes useful, they also keep breaking basic functionality that worked fine before. Also, when they do claim to support new features, sometimes the quality of implementation is so bad that I'd rather just do things the old-fashioned way. I have seen soooo many demonstrations of a site or app that worked fine on Friday evening when the devs signed off on them and then failed embarrassingly on Monday morning in front of clients or management because a browser updated in between.
First, if I have a req for an engineer with a range of $160K-$190K, if you are making $220K I know it's unlikely that you will accept this job.
That seems a dubious assumption, unless the job you are offering has no attractive properties apart from the salary.
If you're concerned about wasting time on applicants who won't be happy with the top end of your potential range, you might do better to just publish the range honestly and save everyone a lot of hassle. If you're just putting some vague statement of compensation like "negotiable" or "depending on experience" on your job ads, you're probably already losing out on some good applicants anyway.
Second, the person who knows you best as an employee is likely your last employer.
And the employee is leaving that employer, which usually means the relationship isn't mutually satisfactory any more. One obvious way that could happen is that the employee hasn't been given the recognition and compensation they deserve in their current position. This doesn't seem a good basis for judging anything about the employee, unless your primary interest is in how easy it will be to take advantage of them while not compensating them fairly.
From the rest of your post, it looks like you are constrained by a relatively rigid system and this is causing other problems for you. There's probably nothing any of us can do to help you with that, but more flexible employers will naturally be able to attract and retain better hires than you can under these conditions. A lot of good developers will come down on the not disclosing side of this question, and unfortunately you're giving an excellent demonstration of why that is.
I didn't see anything referring to any specific location so far in this thread or in the original question, so it seems fair to point out that the person I was replying to before would be way off-base in some places.
In some countries, when you leave a job you get a statement of income to date for the tax year, which you then give to your next employer so they can set up the right deductions on your salary payments. Unless you happen to change jobs around the end of the tax year, you're going to have some tricky questions to answer when your YTD payments only add up to 75% of what they would be with your claimed former salary. (If you're thinking you'll just say you got a raise earlier in the year, remember that the next question is going to be what they did that was so awful that you left so soon after getting such a huge raise.)
Even if your country doesn't have that sort of system, all it takes is one relevant person at the new employer knowing one relevant person who works/worked at the old employer who could give a general idea of the salary range for your previous position, or even one person who recently made the same move as you but gave a very different number for their old salary.
I just told them some made up BS number that was more than I actually made.
Be careful. If you ever lie on a job application, that could come back to haunt you, at any time for as long as you work for that employer, or potentially even later.
For example, depending on where you are and what rights you would normally have as an employee, you might find that an employer who somehow found out about your deception could instantly terminate your employment, but without either the employer or the government being required to offer any of the protections or payments you would normally have enjoyed.
If you're having a really bad day, you might then find that someone who knew about your deception from that employer has moved and is now reviewing your next application somewhere else, with a predictable effect on the result.
It's tempting to just make something up for an obviously loaded question like this, but you're almost always going to be better off giving an uninformative but not incorrect or misleading answer if you can't avoid or redirect the original question.
Leverage is a fantasy. You have no leverage. THEY control who gets hired. Companies will hire whoever will work for the lowest wages with no regard to skill or qualifications.
But conveniently, declining to provide your salary history will often prevent you from accidentally accepting a job with such a poor employer.
There is only one reason a prospective new employer would ever need to know your previous compensation details, and we all know what it is. Asking what level of compensation you're looking for in a new job is a perfectly legitimate question, but there's no reason any honest and reasonable employer can't just ask it directly. If they don't, and if they won't let your refusal to give history go, you might as well take advantage of the insight you've been given at an early stage and walk away before you waste any more of your time on them.
Yes, in a sense I am conflating them. I don't think the categories as you defined them really are independent; while dependencies in some directions are probably much more common than others, I'd say changes in any of those three categories could potentially merit changes in either of the other two to maintain a good overall presentation. Taking that idea a little further, the definitions you gave focus on a single, static view, without reference to how the material is presented in different contexts or in response to different interactions. Those differences can again create dependencies between the three aspects you mentioned that mean you can't always adjust them separately, at least not with good results.
For example, if a chart counts as content in your scheme then where do you draw the line between content and design? Is the choice of colours for the bars in a bar chart also part of the content, or is that crossing into design territory? What about axis marks to show scales, or whether to use a linear or a log scale? What about deciding whether a chart should be a stacked bar or multiple bars?
All of these things are presentation details: they don't affect the underlying information. In that sense they are design elements.
And yet a good choice for any of them may also depend on the information itself: you can't choose a good scale for a chart without knowing the data it shows and in many cases a pattern or anomaly you are trying to illustrate. In that sense, they are content elements.
Now suppose you switched from a "show everything" presentation style on a large screen with plenty of space to some sort of interactive, exploratory style on a smaller screen with an overview displayed to start with and the ability for the viewer to drill down and see more detail selectively. Is this a part of the design in your scheme, or have we now crossed into structure territory? I suggest that the dynamic aspect puts this example beyond any of your original three categories.
In short, while I agree with you that building multiple distinct presentations can be expensive, I don't agree with your implication that isolating the style/skin and the layout and then adapting each of these to different contexts is a good general solution. Sometimes I think you need a more profound change in the presentation style to get decent results, and even when a similar style is sufficient for each relevant environment, sometimes changes in one category would still benefit from corresponding changes in either or both of the others.
I agree with much of what you say, particularly the inappropriateness of trying to shoehorn interfaces from one type of device into another type of device with very different attributes. However, I also think your argument as written is fatally flawed, because content, design and structure aren't really independent.
"A picture is worth a thousand words" is as true as ever. The difference in effectiveness between a table of raw data and a standard but well-chosen chart can be dramatic. Sometimes the difference between a standard chart and an original presentation style designed specifically to illustrate important points about a particular data set can be greater still. That can be something as distinctive as Minard's famous depiction of Napoleon's army campaign into Russia, which Tufte has suggested might be the best statistical graphic ever drawn. It can also be something as "simple" as showing two versions of some text side-by-side and highlighting the differences.
With today's technologies, our presentations can also move and/or respond to user interactions. Sometimes an animated version that shows changes over time can be very useful. Sometimes a UI where you can drill down into more detail in areas of interest or even experiment by modifying some of the data to see the results interactively can highlight important information better than any predetermined presentation. The interaction idea combines just fine with the visualisation/animation ideas if the data fits.
The challenge, in my experience as someone who does a lot of custom UI work, is that sometimes a presentation style that works well on a big screen can't just be swapped out for an equivalent on a smartphone. I've seen some very effective "interactive storytelling" pieces -- the New York Times has published some excellent ones in recent years, for example -- that really do make the most of online reading of articles to do things they could never have done on paper. One that I remember well was an article telling the story of an interesting journey that had been undertaken, and as the article scrolled down as normal on the left half of the page, an animated map on the right half illustrated the progress of the adventurers. When key events were reached in the story, more detail was also added on the map, or sometimes photos or other illustrations would pop up. You couldn't do that in the same style on a 5" smartphone screen, because the side-by-side aspect wouldn't work. Although the underlying information would still be the same, you'd need a different presentation style to make the most of it on a small screen, not just a simple layout change or different typography. For better or worse, that does bring us back to genuinely designed different presentations for different media.
"Mobile first" is partly to blame, but lazy/cheap teams are more so.
Take a look at what's popular in trendy web app design today: flat everything, big rectangular colour blocks, lines and rounded corners, text. Look at the boxy, side-by-side layouts, almost invariably collapsing into increasingly linear formats for narrower screens until it's just a single column.
Now look at what you can do easily and portably with CSS. In particular, look at what you can achieve by just slapping Bootstrap or the like on your site, without spending much time or money considering the design and layout, and certainly without hiring any sort of designer or, $DEITY forbid, a digital artist to create custom graphics that fit the style of your product/service and build any sort of distinctive branding.
There was, at the time, some justification for this in that downloading lots of large images on the mobile networks of a few years ago really could significantly slow down loading a page, with resulting poor user experience and app/site performance. But for most of us, our target markets are on faster networks today, and CDNs are much more developed now as well. And certainly you don't get any allowance for this if your site includes megabytes of JS frameworks, ad content, or auto-playing hero video.
Likewise, there is some justification for minimal UI chrome on small screen devices where every pixel is precious, but you don't get any allowance for this if you replace a simple hairline with half an inch of whitespace because your visual style is so generic and unguided that the user can't actually tell how the UI works otherwise.
Frankly, Microsoft, Google and Apple are amateurs when it comes to nerfing design by being flat and bland. Web developers have been moving in this direction for at least as long as smartphones and tablets have been around, and people with actual UI design skills have been criticising them and pointing out the obvious and horrible usability flaws for just as long.
This is the real problem today, IMHO. No-one really knows yet what the trade arrangements between the UK and the EU will be. That uncertainty is damaging: everyone has to keep their options open and plan for a variety of possible outcomes, which typically isn't free. It's also the reason that the endless doomsaying from certain quarters is not just unhelpful but potentially dangerous: it's far too early to know whether any specific prediction will actually happen, but if enough people repeating claims about terrible consequences anyway, it will eventually damage sentiment and potentially turn some of those consequences into self-fulfilling prophecies.
The trouble with your argument is that it is essentially saying "what the voters actually wanted doesn't make sense, but let's do something".
The thing is, I have yet to meet a voter (as far as I'm aware) who wanted any of the usual stereotype positions ascribed to entire groups of voters by the media and/or strident believers in some other point of view.
The vast majority of people I've spoken to about Brexit basically wanted to maintain some degree of ties with the EU, but didn't want to follow the "ever closer union" direction it's going in. Some ultimately came down on the side of Remain, hoping that the situation can be fixed from within and trusting that the UK's various exemptions would protect us in the meantime, or feeling that the short term damage and/or general uncertainty about leaving wasn't worth the risk. Some ultimately came down on the side of Leave, hoping that some alternative arrangement would suit all parties, but feeling that the limitations on being able to forge our own path if we remained in were too great and/or that it wasn't realistic to remain a member while claiming ever more special rules just for us. In some cases, views on the likely future of the EU itself also played a part, either seeing it as a positive influence for say preventing wars, or as a risk say because the underlying European economic situation still doesn't look good.
I don't see anything nonsensical about any of those positions. They're just weighing up various factors, all of which would probably be true to some extent, but reaching different conclusions about how they balance.
However, none of those people were hardcore Europhiles who thought there were no potential benefits to leaving, xenophobic nationalists who thought all immigrants were evil and EU membership had no benefits, or that weird in-the-middle group who apparently think we should have Brexit but then somehow stay in all the parts of the EU that actually affect us anyway. To me, those three positions are all logically inconsistent with the evidence and/or themselves, but the way a lot of the reporting (and, frankly, a lot of the comments in online forums) have been going, you'd think they were the only options.
You think manufacturing exporters who just got a 10-20% windfall through the exchange rate changes are going to be scared about the risk of WTO tariffs that will be an order of magnitude smaller in most sectors? It still seems like you're overstating the case here.
I'm guessing the main concern among the listed methods is fingerprints. Unlike some other biometric identifiers, fingerprints linger, and they get picked up at places like crime scenes. Given the questionable standards of forensic analysis in criminal cases in recent years, the potential fishing expeditions when fingerprints are found in connection with serious crimes, and the scary potential consequences if you're involved in a case of mistaken identity, I can entirely understand why some people would be hesitant about giving any government their fingerprints (or a DNA sample, for the same reasons and more).
Something like an iris scan seems significantly less problematic from that point of view. It's still a useful identifier for practical purposes, but it lacks the persistence of fingerprints or DNA, it lacks the ability to identify covertly at a long distance like a voiceprint, facial recognition or gait analysis, and crucially, it will probably continue to lack those risks for a considerable time, because physics.
I'm not saying there aren't real concerns there, but let's keep this in perspective. The UK economy is heavily service-based. A large majority of the products made and services supplied in the UK are not exported to customers in the EU; most stay within the UK, and while trade with EU partners is obviously very important, the majority of the UK's international trade is with non-EU partners. And even if trade with the EU and partners with existing trade deals via the EU falls back on WTO rules, the economists being interviewed by the serious news organisations seem to be saying it's highly sector-specific but overall you would expect tariffs to have roughly a 2% effect. That's significant, but it's far less than the damage caused by say the credit crunch, and it's far less than the fluctuations due to exchange rates in recent months.
Obviously it's sensible for international businesses to look at their situation as their environment changes and to adapt accordingly. I don't see how anyone could reasonably criticise them for that. But at the same time there's definitely a bit of opportunist smoke blowing going on, and there are definitely big businesses angling for some sort of sweetheart deal with the UK government here. I don't think it makes much sense to argue the UK has closed up shop when we literally haven't even started the real Brexit process yet and there's still considerable political negotiation to come both at home and with the remaining EU members.
Slashdot Mirror
I don't know enough details about the modern Linux kernel to comment usefully on that specific case, but certainly as an industry we have a lot more experience with hot reloading techniques today than we had a few years ago, and certainly a lot of devices don't have or need anything like the complexity of modern desktop/server operating systems.
I deliberately wrote a figurative 99% of the time rather than 100% before, because in the real world you might still have a situation where you really can't update the very heart of a system without taking it out of service at least briefly. Even with perfect software design, sometimes the hardware limitations might get you. But with reasonably designed, modular software architecture, we really shouldn't see devices becoming useless for extended periods anything like as often as we do in practice at the moment.
I don't have any special insider information or anything. I just looked up the most recent official data I could find for Adobe, and highlighted a few key points that stood out to me for comparison.
Since you asked, my guess is that there's a simple explanation for Office 365, something like the following.
Firstly, Office 365 is on sale at the same time as a more traditional version in Office 2016. It's reasonable to assume that some people and organisations prefer the subscription/rental model, while others prefer a more traditional permanent purchase.
Now, some of the pro-subscription group have already subscribed, so that leaves fewer new subscriptions from among that group now (and more renewals). Also, some of the pro-purchase group will presumably have updated to 2016 recently, so they may be less inclined to throw that investment out by signing up for 365 so soon. Both of these point to a drop in the number of new subscriptions.
To reverse that trend, Microsoft would need to find more people who want to have Office and/or get more existing customers interested in subscribing instead of purchasing outright. However, unlike say Adobe or Autodesk, Microsoft can't easily force their existing customers' hands by going subscription-only. It's a relatively mature and saturated market. Existing versions already do most of what most people need in an Office-style product, so few people switch or upgrade once they've got their copy. There's more innovation in related products nipping at the edges of the market or aiming at specific niches, but O365 isn't really addressing that part of the industry. And of course the elephant in the room is that there is credible, direct competition to Office, so if Microsoft tried to force happy, purchase-oriented customers to move, they might find Office 365 was not where those customers moved to.
So, as long as they want to sit on their 800lb gorilla cash cow without making a game-changing new release, MS might just have to accept that a lot of customers still value Office enough to buy new copies but many don't want to rent. If so then, as our showbiz friends might say, they need to give the people what they want. That inevitably leads to some slowing down of new subscriptions as the market reaches an equilibrium between the two models.
As I said, I have no special insider knowledge to back this up. I'm just applying Occam's Razor.
Here's an investor presentation from last week (PDF). It's a long document, but the following is mostly derived from a couple of slides at the bottom of page 3.
It looks like their total revenues for Creative Cloud dipped a few percent and then recovered again over the period 2012-2015, and as of 2016 their annual recurring revenue for that area is up to around $3.5B, compared to annual revenue of around $2.5B back in 2012 when their subscription model was starting up.
Over the same four-year window, it appears that their subscription ARR has been increasing roughly linearly, while their non-subscription revenues are fast approaching zero.
In short, it looks like they are now better off than they were four years ago in terms of annual Creative Cloud revenue, by about 40% if they maintain their current subscription level.
Another figure they mention is current year-on-year subscription growth of 46% outside the US. However, they are deafeningly quiet on what proportion of their overall market that represents or the equivalent figure for US customers. Their overall growth rate is clearly far less than that, so it could be that they're successfully expanding into foreign markets and that's helping to drive their overall subscription growth (probably a good thing for Adobe) but it could also be that sales in foreign markets are covering up a significant reduction in the US as increasing numbers of US customers are cancelling their subscriptions (probably a bad thing for Adobe).
It's also difficult to tell how many subscribers they actually have, since there doesn't seem to be any breakdown of which of the available subscription plans are generating how much revenue or what sort of effects they see from volume licensing, subscribers from different countries, or subscribers paying in different currencies. If we guess an average subscriber is worth about US$500 per year to them in revenues, that would give them around 7 million current subscribers, but this could obviously be way off if say most of the revenues are actually from enterprise customers paying far less than the headline per-seat prices with their volume deals.
Obviously the devil is in the details but NFC keycards aren't going anywhere (no changing locks and lost keys)
OK, I'm with you so far.
and internet aware locks are the obvious next step of convenience and cost cutting.
::boggle::
Even some of the cheaper hotel chains here in the UK now routinely have machines that let you check in without staff intervention, including coding your keycards for you. It takes a few moments. It is not at all obvious to me that Internet-enabling anything about this process would be either more convenient or cheaper for anyone.
Maybe. The article doesn't say, but if the attack truly did lock the doors from inside then clearly they have other problems, for sure. At that point, you're not talking about vulnerabilities to hostile parties any more, you're talking about basic reliability and safety concerns.
Well, yes. People with physical access to closed systems can potentially attack them in ways that people with only remote access can't. News at 11.
That's not an argument against minimizing the attack surface by avoiding unnecessary remote access, though, is it?
Obviously there should be physical safeguards for when the tech screws up, but I don't think that diminishes the scale of the original screw up.
The article doesn't specify the exact system or how it was compromised, so unless you have some other source to share, none of us know whether the devices that were compromised in this specific case were directly Internet-connected. Some modern hotel systems are. It could also be that the repeated hacks in this case accessed the room key system indirectly via some other system that was compromised first. The fundamental issues raised are the same either way.
Who thought it was a good idea for essential systems like this to be online in the first place?!
This is why the Internet of Things is such a horrible concept. Most things don't need to be online and connected to everything else, and the cost of trying to be trendy is huge increases in risks to the privacy, security and reliability of everyday items.
Closed networks do just fine for these kinds of systems, don't actually need to cost that much more, and have none of the vulnerabilities.
But as a web developer, Google Chrome and the other "evergreen" browsers are a nightmare for stability. While having support for newer HTML or CSS or whatever is sometimes useful, they also keep breaking basic functionality that worked fine before. Also, when they do claim to support new features, sometimes the quality of implementation is so bad that I'd rather just do things the old-fashioned way. I have seen soooo many demonstrations of a site or app that worked fine on Friday evening when the devs signed off on them and then failed embarrassingly on Monday morning in front of clients or management because a browser updated in between.
It's also reasonable to ask why so many updates need to reboot the whole device these days.
99% of the time, that in itself reflects a weakness in the underlying OS and software architecture.
First, if I have a req for an engineer with a range of $160K-$190K, if you are making $220K I know it's unlikely that you will accept this job.
That seems a dubious assumption, unless the job you are offering has no attractive properties apart from the salary.
If you're concerned about wasting time on applicants who won't be happy with the top end of your potential range, you might do better to just publish the range honestly and save everyone a lot of hassle. If you're just putting some vague statement of compensation like "negotiable" or "depending on experience" on your job ads, you're probably already losing out on some good applicants anyway.
Second, the person who knows you best as an employee is likely your last employer.
And the employee is leaving that employer, which usually means the relationship isn't mutually satisfactory any more. One obvious way that could happen is that the employee hasn't been given the recognition and compensation they deserve in their current position. This doesn't seem a good basis for judging anything about the employee, unless your primary interest is in how easy it will be to take advantage of them while not compensating them fairly.
From the rest of your post, it looks like you are constrained by a relatively rigid system and this is causing other problems for you. There's probably nothing any of us can do to help you with that, but more flexible employers will naturally be able to attract and retain better hires than you can under these conditions. A lot of good developers will come down on the not disclosing side of this question, and unfortunately you're giving an excellent demonstration of why that is.
Well, OK, but most of the world is not the US. :-)
I didn't see anything referring to any specific location so far in this thread or in the original question, so it seems fair to point out that the person I was replying to before would be way off-base in some places.
You're kidding, right?
In some countries, when you leave a job you get a statement of income to date for the tax year, which you then give to your next employer so they can set up the right deductions on your salary payments. Unless you happen to change jobs around the end of the tax year, you're going to have some tricky questions to answer when your YTD payments only add up to 75% of what they would be with your claimed former salary. (If you're thinking you'll just say you got a raise earlier in the year, remember that the next question is going to be what they did that was so awful that you left so soon after getting such a huge raise.)
Even if your country doesn't have that sort of system, all it takes is one relevant person at the new employer knowing one relevant person who works/worked at the old employer who could give a general idea of the salary range for your previous position, or even one person who recently made the same move as you but gave a very different number for their old salary.
I just told them some made up BS number that was more than I actually made.
Be careful. If you ever lie on a job application, that could come back to haunt you, at any time for as long as you work for that employer, or potentially even later.
For example, depending on where you are and what rights you would normally have as an employee, you might find that an employer who somehow found out about your deception could instantly terminate your employment, but without either the employer or the government being required to offer any of the protections or payments you would normally have enjoyed.
If you're having a really bad day, you might then find that someone who knew about your deception from that employer has moved and is now reviewing your next application somewhere else, with a predictable effect on the result.
It's tempting to just make something up for an obviously loaded question like this, but you're almost always going to be better off giving an uninformative but not incorrect or misleading answer if you can't avoid or redirect the original question.
Leverage is a fantasy. You have no leverage. THEY control who gets hired. Companies will hire whoever will work for the lowest wages with no regard to skill or qualifications.
But conveniently, declining to provide your salary history will often prevent you from accidentally accepting a job with such a poor employer.
There is only one reason a prospective new employer would ever need to know your previous compensation details, and we all know what it is. Asking what level of compensation you're looking for in a new job is a perfectly legitimate question, but there's no reason any honest and reasonable employer can't just ask it directly. If they don't, and if they won't let your refusal to give history go, you might as well take advantage of the insight you've been given at an early stage and walk away before you waste any more of your time on them.
Yes, in a sense I am conflating them. I don't think the categories as you defined them really are independent; while dependencies in some directions are probably much more common than others, I'd say changes in any of those three categories could potentially merit changes in either of the other two to maintain a good overall presentation. Taking that idea a little further, the definitions you gave focus on a single, static view, without reference to how the material is presented in different contexts or in response to different interactions. Those differences can again create dependencies between the three aspects you mentioned that mean you can't always adjust them separately, at least not with good results.
For example, if a chart counts as content in your scheme then where do you draw the line between content and design? Is the choice of colours for the bars in a bar chart also part of the content, or is that crossing into design territory? What about axis marks to show scales, or whether to use a linear or a log scale? What about deciding whether a chart should be a stacked bar or multiple bars?
All of these things are presentation details: they don't affect the underlying information. In that sense they are design elements.
And yet a good choice for any of them may also depend on the information itself: you can't choose a good scale for a chart without knowing the data it shows and in many cases a pattern or anomaly you are trying to illustrate. In that sense, they are content elements.
Now suppose you switched from a "show everything" presentation style on a large screen with plenty of space to some sort of interactive, exploratory style on a smaller screen with an overview displayed to start with and the ability for the viewer to drill down and see more detail selectively. Is this a part of the design in your scheme, or have we now crossed into structure territory? I suggest that the dynamic aspect puts this example beyond any of your original three categories.
In short, while I agree with you that building multiple distinct presentations can be expensive, I don't agree with your implication that isolating the style/skin and the layout and then adapting each of these to different contexts is a good general solution. Sometimes I think you need a more profound change in the presentation style to get decent results, and even when a similar style is sufficient for each relevant environment, sometimes changes in one category would still benefit from corresponding changes in either or both of the others.
I agree with much of what you say, particularly the inappropriateness of trying to shoehorn interfaces from one type of device into another type of device with very different attributes. However, I also think your argument as written is fatally flawed, because content, design and structure aren't really independent.
"A picture is worth a thousand words" is as true as ever. The difference in effectiveness between a table of raw data and a standard but well-chosen chart can be dramatic. Sometimes the difference between a standard chart and an original presentation style designed specifically to illustrate important points about a particular data set can be greater still. That can be something as distinctive as Minard's famous depiction of Napoleon's army campaign into Russia, which Tufte has suggested might be the best statistical graphic ever drawn. It can also be something as "simple" as showing two versions of some text side-by-side and highlighting the differences.
With today's technologies, our presentations can also move and/or respond to user interactions. Sometimes an animated version that shows changes over time can be very useful. Sometimes a UI where you can drill down into more detail in areas of interest or even experiment by modifying some of the data to see the results interactively can highlight important information better than any predetermined presentation. The interaction idea combines just fine with the visualisation/animation ideas if the data fits.
The challenge, in my experience as someone who does a lot of custom UI work, is that sometimes a presentation style that works well on a big screen can't just be swapped out for an equivalent on a smartphone. I've seen some very effective "interactive storytelling" pieces -- the New York Times has published some excellent ones in recent years, for example -- that really do make the most of online reading of articles to do things they could never have done on paper. One that I remember well was an article telling the story of an interesting journey that had been undertaken, and as the article scrolled down as normal on the left half of the page, an animated map on the right half illustrated the progress of the adventurers. When key events were reached in the story, more detail was also added on the map, or sometimes photos or other illustrations would pop up. You couldn't do that in the same style on a 5" smartphone screen, because the side-by-side aspect wouldn't work. Although the underlying information would still be the same, you'd need a different presentation style to make the most of it on a small screen, not just a simple layout change or different typography. For better or worse, that does bring us back to genuinely designed different presentations for different media.
"Mobile first" is partly to blame, but lazy/cheap teams are more so.
Take a look at what's popular in trendy web app design today: flat everything, big rectangular colour blocks, lines and rounded corners, text. Look at the boxy, side-by-side layouts, almost invariably collapsing into increasingly linear formats for narrower screens until it's just a single column.
Now look at what you can do easily and portably with CSS. In particular, look at what you can achieve by just slapping Bootstrap or the like on your site, without spending much time or money considering the design and layout, and certainly without hiring any sort of designer or, $DEITY forbid, a digital artist to create custom graphics that fit the style of your product/service and build any sort of distinctive branding.
There was, at the time, some justification for this in that downloading lots of large images on the mobile networks of a few years ago really could significantly slow down loading a page, with resulting poor user experience and app/site performance. But for most of us, our target markets are on faster networks today, and CDNs are much more developed now as well. And certainly you don't get any allowance for this if your site includes megabytes of JS frameworks, ad content, or auto-playing hero video.
Likewise, there is some justification for minimal UI chrome on small screen devices where every pixel is precious, but you don't get any allowance for this if you replace a simple hairline with half an inch of whitespace because your visual style is so generic and unguided that the user can't actually tell how the UI works otherwise.
Frankly, Microsoft, Google and Apple are amateurs when it comes to nerfing design by being flat and bland. Web developers have been moving in this direction for at least as long as smartphones and tablets have been around, and people with actual UI design skills have been criticising them and pointing out the obvious and horrible usability flaws for just as long.
This is the real problem today, IMHO. No-one really knows yet what the trade arrangements between the UK and the EU will be. That uncertainty is damaging: everyone has to keep their options open and plan for a variety of possible outcomes, which typically isn't free. It's also the reason that the endless doomsaying from certain quarters is not just unhelpful but potentially dangerous: it's far too early to know whether any specific prediction will actually happen, but if enough people repeating claims about terrible consequences anyway, it will eventually damage sentiment and potentially turn some of those consequences into self-fulfilling prophecies.
Sorry, you're quite right, it was retina scanning that I was thinking of.
The trouble with your argument is that it is essentially saying "what the voters actually wanted doesn't make sense, but let's do something".
The thing is, I have yet to meet a voter (as far as I'm aware) who wanted any of the usual stereotype positions ascribed to entire groups of voters by the media and/or strident believers in some other point of view.
The vast majority of people I've spoken to about Brexit basically wanted to maintain some degree of ties with the EU, but didn't want to follow the "ever closer union" direction it's going in. Some ultimately came down on the side of Remain, hoping that the situation can be fixed from within and trusting that the UK's various exemptions would protect us in the meantime, or feeling that the short term damage and/or general uncertainty about leaving wasn't worth the risk. Some ultimately came down on the side of Leave, hoping that some alternative arrangement would suit all parties, but feeling that the limitations on being able to forge our own path if we remained in were too great and/or that it wasn't realistic to remain a member while claiming ever more special rules just for us. In some cases, views on the likely future of the EU itself also played a part, either seeing it as a positive influence for say preventing wars, or as a risk say because the underlying European economic situation still doesn't look good.
I don't see anything nonsensical about any of those positions. They're just weighing up various factors, all of which would probably be true to some extent, but reaching different conclusions about how they balance.
However, none of those people were hardcore Europhiles who thought there were no potential benefits to leaving, xenophobic nationalists who thought all immigrants were evil and EU membership had no benefits, or that weird in-the-middle group who apparently think we should have Brexit but then somehow stay in all the parts of the EU that actually affect us anyway. To me, those three positions are all logically inconsistent with the evidence and/or themselves, but the way a lot of the reporting (and, frankly, a lot of the comments in online forums) have been going, you'd think they were the only options.
You think manufacturing exporters who just got a 10-20% windfall through the exchange rate changes are going to be scared about the risk of WTO tariffs that will be an order of magnitude smaller in most sectors? It still seems like you're overstating the case here.
I'm guessing the main concern among the listed methods is fingerprints. Unlike some other biometric identifiers, fingerprints linger, and they get picked up at places like crime scenes. Given the questionable standards of forensic analysis in criminal cases in recent years, the potential fishing expeditions when fingerprints are found in connection with serious crimes, and the scary potential consequences if you're involved in a case of mistaken identity, I can entirely understand why some people would be hesitant about giving any government their fingerprints (or a DNA sample, for the same reasons and more).
Something like an iris scan seems significantly less problematic from that point of view. It's still a useful identifier for practical purposes, but it lacks the persistence of fingerprints or DNA, it lacks the ability to identify covertly at a long distance like a voiceprint, facial recognition or gait analysis, and crucially, it will probably continue to lack those risks for a considerable time, because physics.
I'm not saying there aren't real concerns there, but let's keep this in perspective. The UK economy is heavily service-based. A large majority of the products made and services supplied in the UK are not exported to customers in the EU; most stay within the UK, and while trade with EU partners is obviously very important, the majority of the UK's international trade is with non-EU partners. And even if trade with the EU and partners with existing trade deals via the EU falls back on WTO rules, the economists being interviewed by the serious news organisations seem to be saying it's highly sector-specific but overall you would expect tariffs to have roughly a 2% effect. That's significant, but it's far less than the damage caused by say the credit crunch, and it's far less than the fluctuations due to exchange rates in recent months.
Obviously it's sensible for international businesses to look at their situation as their environment changes and to adapt accordingly. I don't see how anyone could reasonably criticise them for that. But at the same time there's definitely a bit of opportunist smoke blowing going on, and there are definitely big businesses angling for some sort of sweetheart deal with the UK government here. I don't think it makes much sense to argue the UK has closed up shop when we literally haven't even started the real Brexit process yet and there's still considerable political negotiation to come both at home and with the remaining EU members.