Our software is perfect. If something has gone wrong, it must be YOUR fault "That's it! I've had enough of this! I'm leaving!" How tiresome. Oh well, let them leave. Perhaps then we'll get some peace – although I doubt it. Sure enough, a quick scroll downwards reveals three more posts from users that detail how a small piece of commercial software fails to meet their high standards. Each whiney post …
Not sure that any UX designer nowadays even knows how to make a function UX - most of them seem to be heading towards making everything the same colour, having no consistency between different parts of the app, and my personal favourite the pop-under window. They mostly look shit as well.
I think the problem also might be that the management often thinks that Full Stack also means UX. So they make things as they go along. "If it's easy for me to use, then anyone should be able to use it."
Then the software turns into a mess and they decide to hire a dedicated UX - which has quite an impossible job:
UX: "Here is the new optimised flow, that would improve usability with the least possible amount of changes"
Devs: "Oh we can't do that, we would have to rewrite that part of application."
Manager: "Can we at least change colours?"
UX *please kill me*
I've been on the other side of this, as the developer of a large system when a UX designer was called in.
The UX designer couldn't come in-house to talk with the devs, and the scope he was given by senior management only ever grew.
By the time he was finished, he had produced a design for a new section of the website that would need a Google-sized team to pull off. Can't blame him - it looked good and it was what he was asked to design. But there was no way in hell I'd ever have the time or resource to build it.
Sometimes it's the simplest things that they get wrong: for an example, a timesheet data entry page which requires the start and end times to be in hh:mm format, but which not only does not automatically calculate and populate an hours worked field, requires it - without comment or cue - to be filled in in decimal hours.
Which reminds me, I haven't submitted myself to it this week yet.
That's similar to the brain dead "developers" who are so incompetent that they are unable to parse the spaces out of a credit card number. This is the number that on almost every card is presented as four sets of four digits with spaces between for ease of human use must be entered without spaces.
This post has been deleted by its author
Not all postal codes and phone numbers in the world use the Merkan formats.
For years I have quoted my phone number in standard international format. It starts +44, and all the numbers remembered by my mobile phone follow the same rule. That means that the numbers work wherever in the world I can get service; the operators have known this since at least the introduction of GSM.
So why can't web developers get their brains in gear?
Only this morning I was filling in an online form. My browser helpfully allows various boxes to be filled in with one click. The box simply labelled "Telephone number" gets filled in with that +44 number.
On clicking their "OK" button I am informed that the phone number "MUST BE IN THE FORMAT SPECIFIED". The format they require is still a mystery to me. Does it demand spaces? Hyphens? Round brackets?
They didn't get my business.
The worst part of that is that there IS a standard, I have been using that standard format for decades.
To be honest, I find it very annoying if a phone entry field doesn't comply with that format, but I'm also not a big fan of the country code lookup most have attached - I am 30x quicker just typing the country code.
Addresses are the worse. In the US I assume that everyone has address that's street, town, state, zip, but the UK needs way more lines. Clients I deal with have Office Name, Building Name/No, Industrial Estate Name, Street, Town. Postcode. That's four items to fit in the single field for "Street".
… why does the Contacts app on my Mac only allow one line for street address? I'd assumed that US addresses were uniform across the country.
The obvious answer is, in this case, the correct answer, but since you’d asked that it be set aside, it won’t be mentioned again.
US addresses are largely, but not entirely, uniform across the country. A domestic US business address can potentially contain up to eleven lines, but typically five lines will suffice. A domestic US residential address typically has three lines (recipient; delivery address; city, state/commonwealth/territory, postal code), but a long delivery address can stretch over two lines, with the line containing the final part of the delivery address coming directly before the line containing its initial part. An optional “attention” line can come before the recipient line. Similarly, an optional private mail stop code line can come before the attention line (if present) or the recipient line (if no attention line is present). Dual addressing (where the physical address differs from the delivery address, e.g. with a house as the former and a post office box as the latter) is possible, but not recommended; in this case, the physical address is put on a separate line directly before the delivery address.
In the case of Puerto Rico, condominium names and urbanización (a housing estate area) names can appear on a separate line directly before the delivery address. Many condos and urbanizaciones there lack street addresses, though, so in those cases, those names function as delivery addresses.
"recipient; delivery address; city..."
Just stop there. That gets carried over to the UK. A lot of UK citizens may live in a city but many don't.
Then there's an assumption that every house has a number. I've run into that trying to order stuff online where with a site refusing to accept an address without one.
A lot of UK citizens may live in a city but many don't.
“City” is common shorthand in addresses here for “city, borough, town, village, military base, or unincorporated area”, but technically, the “city” in a US address really represents a US post office. Every domestic US address has a “city” because every piece of first-class mail is delivered to or through the US post office in the destination address. Packages can be delivered by private delivery services, but they tend to follow the US postal service addressing standards, even if no US post office is involved in a package’s delivery.
Then there's an assumption that every house has a number.
Corner addresses are accepted by the postal service here (e.g. “4th and Main”), but street addresses are preferred (e.g. “404 Main Street”). Some addresses here (e.g. on rural routes and highway contract routes) lack both streets and house numbers, instead having route numbers and box numbers (or unit numbers and box numbers in the case of military addresses); for the former, I’d imagine that private delivery services would need a street name and house number instead. It could well be, though, that every US address that is specified with exactly one named/numbered street has an associated house (well, building, really) number.
"in a US address really represents a US post office"
In the UK "city" has a very specific legal meaning in that it denotes somewhere with a royal charter. It doesn't even have to be a very big place.
In Royal Mail parlance - or at least in terms of the field name in PAF - the equivalent term would be "Post town". I don't expect the US to actually use UK terminology but they could at least think that their S/W might be used where their norms don't apply and try for a neutral term.
But it's worse than that. I've seen it used in genealogical S/W where the historical context could well pre-date the US postal service so even for a US location it would be an anachronism. It's not adherence to some standard, it's just lack of thought.
As to street numbers "preferred" doesn't mean anything. S/W needs to be able to accept the actual format of the real data without preferences or assumptions. If you can't do that you run into trouble. I had major problem getting a parcel delivered from the German vendor whose site didn't accept non-numeric addresses, not helped by DPD insisting that my address is at some single location at the geographic centre of the post code and I get the impression that their drivers aren't even allowed to deviate from that even if it's wrong.
In the UK “city” has a very specific legal meaning in that it denotes somewhere with a royal charter. It doesn't even have to be a very big place.
Except for the word “royal”, the same applies in the US. However, the legal meaning is not the meaning that is used in the US with postal addresses.
I don't expect the US to actually use UK terminology but they could at least think that their S/W might be used where their norms don’t apply and try for a neutral term.
The most accurate neutral term for the “post town” component of addresses here would be “post office”, but “city” is too well entrenched in common usage here to be replaced. “City” also has the advantage of brevity over “post office” (and “post town”).
I’ve seen it used in genealogical S/W where the historical context could well pre-date the US postal service so even for a US location it would be an anachronism. It’s not adherence to some standard, it’s just lack of thought.
The legal definition of “city” is not the only definition of “city”. Note that the first definition of “city” in the OED is
A town or other inhabited place. Not a native designation, but app. at first a somewhat grandiose title, used instead of the OE. burh, BOROUGH. Frequently applied (after civitas of the Vulgate = πόλις of N.T. and LXX.) to places mentioned in the Bible which were really mere villages, e.g. Nazareth, Nain, Bethlehem; here, as a literalism of translation, it still stands in Bible versions.
(My bolding above.)
As to street numbers “preferred” doesn’t mean anything. S/W needs to be able to accept the actual format of the real data without preferences or assumptions. If you can’t do that you run into trouble.
I agree with your view of what software should do. My mention of “preferred” (and “not recommended” in reference to dual addressing) was purely in reference to domestic US postal processing of address formats. A letter or parcel will usually get to its destination whether an address is in a “preferred” form or a “not recommended” form, but the speed of delivery could be affected.
Much of modern software, however, seems to suffer from the “bias for action” model of development — “if we didn’t get it right the first time around, we’ll hear about it soon enough and ship a quick fix to the problem”. (See Timochka’s comment for an unfortunately common example.) To paraphrase an old slogan for Zenith televisions, “the quality goes in before the name falls off”.
It depends on the application, I think. If you need information for a delivery or a database that needs to differentiate names, titles, locations and so on then I think you're correct - although 4 lines for address isn't enough for many of my contacts. If you just want to record someones address then an unlimited text box is best for me. That's how I've come to use Apples Contacts app on the Mac. In spite of what Apple and many posters-in-forums claim, I haven't found a reliable way to paste an address into Mac's contacts without doing it line-by-line so now I just paste the address into the notes field. It buggers up my Christmas-card label automation, but now it's set up it's just as easy to open Libre office and paste a new address into the labels file as it is to parse and paste the different bits of the address between the limited lines available in Contacts then add multiple commas to split them in the merge, then run the mail merge, and lose my temper a few times along the way.
It was all supposed to be so eeeeaaaasy!
This post has been deleted by its author
My fathers official address:
House Name
Postal Town (absolutely miles away)
Post Code
i.e. no street name, no town name - so many UX cannot deal with that we have invented a street and town name! If the UX does a postcode search, we are buggered!
Worse, his business telephone number stopped appearing in the local phone books, and he only noticed when customers started asking when he had stopped trading. His STD (telephone) number was for the local town, and yet the UK telephone books moved over to Postal Town based, which was many miles away. No amount of complaining got him back into the local town telephone book that the STD code was for.
I have friends in a village in Leicestershire and yet their postal town is in Derbyshire - that is terribly confusing as well.
I have friends in a village in Leicestershire and yet their postal town is in Derbyshire - that is terribly confusing as well.
The post codes for Bedford, the county town of Bedfordshire, all have Milton Keynes out-codes (MK). Milton Keynes is in Buckinghamshire, and actually quite a way from Bedford.
Luton, on the other hand, the other large town in Bedfordshire, has its own postcode out-code (LU).
Of course, Bedford is a shit-hole and deserves no recognition, and I can say that, having had the misfortune of spending my formative years there.
I'd assumed that US addresses were uniform across the country.
No, they are not.
For one thing, street addresses are not always mailing addresses. The USPS does not deliver to every physical address. So forms which ask for an address and fail to specify street or mailing are already wrong for the US (and for probably the rest of the world as well). On a related note, vendors who don't specify how they ship are Doing It Wrong, because USPS and private delivery services such as UPS and FedEx have different requirements, so the address you have to provide may depend on which carrier they use.
For either type of address, private residences may have fractional building numbers, building "numbers" that include letters, and so forth. Institutional addresses may have building names or codes.
People may want to use five-digit, nine-digit, or fourteen-digit zip codes. They may need to add an attention line or other delivery instructions.
The simple fact of the matter is that web-form designers make all sorts of assumptions about addressing, and they're rarely called out on it.
We get similar dumbness in the UK when County is a mandatory postal address field. It isn't, it's not been a required part of addresses since the 1990s when the Post Office probably got bored of all the gerrymandering going on and changes to counties so just removed it from postal addresses. We still get idiot developers writing systems requiring Country for a postal address, sometimes implementing this stupidity as a drop down list. Which is all great except it's not required and fails for large cities such as London that don't have a county.
It gets worse where US-based companies think they know better than the locals how UK addresses work.
Ancestry, for example, uses a gazetteer which has little basis in reality. According to them, as a sample, Tower Hamlets is in Kent. This sort of insanity means that their search function will ALWAYS fail to locate records which include the correct county.
Other sites insist that historical events took place in places with modern-day administrative names. Wikipedia, for example, insist that George Washington was born in the United States of America, which I am sure would have been news to him.
"Wikipedia, for example, insist that George Washington was born in the United States of America, which I am sure would have been news to him."
No, they don't. Here's the sentence discussing his place of birth:
George Washington was born on February 22, 1732,[b] at Popes Creek in Westmoreland County, in the British colony of Virginia,[17] and was the first of six children of Augustine and Mary Ball Washington.
And here's the short summary in the table:
Born
February 22, 1732 [O.S. February 11, 1731]
Popes Creek, Virginia, British America
Died
December 14, 1799 (aged 67)
Mount Vernon, Virginia, U.S.
similar dumbness in the UK when County is a mandatory postal address field
We have three main problems with dumb forms, and the county is two of them. There has been no such county as "Mid Glamorgan" since 1996 or so, which probably predates most of these databases, yet some of them insist on adding "Mid Glamorgan" once you enter the postcode.
Our own postal town also gives its name to the new Unitary Authority (the administrative areas which replaced counties in 1996) and since many forms insist on both "town" and "county" you have to fill the same thing in twice. I've mentioned this here before, but many years ago, when auto-fill address databases were just becoming a thing, one form I had to fill in didn't have our address at all, so I put the English spelling for the town and the Welsh spelling for the county (either is acceptable by the way) and to this day I occasionally come across an auto-fill form which auto fills with those two spellings.
A slightly unrelated problem we have is that our house doesn't exist, according to at least one database. We made the mistake of de-registering the address while we were rebuilding. It saved us 18 months or so of Council Tax* but within 6 months the address had disappeared from databases. We re-registered the property nearly two years ago now, yet at least one database still only knows about numbers 1, 2, 3, 5, 6, 7, 8 and 9 in the street! Fortunately it still allows you to enter the address manually.
M.
*warning for anyone else considering this, well in Wales anyway. Upon re-registering the property is re-valued, in much the same way as would happen on sale (I gather this doesn't happen in the same way in England) and we were bumped up a band because we added a bedroom. If we had been able to continue paying Council Tax for the 24 months we were out of the property** it would not have been re-valued and we would still be paying the old rate. It'll take a few years, but in time we will have paid more because of the higher banding than we saved by not paying for 18 months :-(
**you can get a 6-month Council Tax "holiday" under this sort of circumstance, which is probably enough for many renovation jobs, but ours turned into a bit of a long-winded affair for reasons people are getting bored with hearing me explain by now :-)
In Bristol, you variously end up with no county (technically wrong), Bristol as the city and the county (technically correct, but redundant in an address), Avon as the county (hasn't existed since 1996), or a neighbouring county, such as, variously, Somerset, North Somerset, Gloucestershire, or South Gloucestershire, all of which are wrong.
Normally when filling badly designed forms that require a county, I put Bristol in as the county, and the area of Bristol I live in as the town. If the county is a drop-down and doesn't include Bristol, you may find it does include Avon, even though that county hasn't existed since before web-forms were really a thing.
Go figure.
> when County is a mandatory postal address field
Often backed up with a drop-down list which is woefully out of date, like by years. I live in one of the counties which succeeded Avon (UK), when that administrative unit was dissolved... in 1996. You wouldn't believe how many drop-downs don't have my actual county's name, have Avon instead, and insist that I use the latter.
Edit: I see that compatriots have made this very point below.
Central European addresses are completely different from anywhere else. In small villages there are no street names, just a seemingly random house number (mine is 125 and across the street is 100). My Canadian bank also doesn’t accept my 12 digit European phone number or my 5 digit postal code as valid.
My mother's house in Scotland had the address '1/2 [name of village]', and an Inverness post code. Inverness is a three hour drive away.
Meanwhile in this village near Berlin (the original), some but not all the houses on the main street carry two numbers, apparently as a result of some historical renumbering exercise. The offset between the old and new numbers is not constant, and some carry only one number. I guess the postman knows where Frau Schmidt lives.
I'd say more like corporations not willing to pay for having spaces parsed out.
You can't expect a poor soul on a wage close to unskilled one, to work their bottom off.
It's not also something a developer just decides to do. It's starting from a product person, then user researcher, ux designer and so on.
Then meetings about spaces can easily take half a day and one has to actually implement it, test, which adds a day easily and delivery manager may simply say "let's leave that for post MVP".
Calling a developer "brain dead" is like calling front line staff same thing when they don't actually make decisions.
"brain dead" is quite accurate I'm afraid.
I've had responses such as the following:
Then only reason for that I can think is - you pay peanuts you get monkeys.
If developer does not have a motivation to go extra mile, then why should they?
It's like buying a slice of toast with butter and complaining the server didn't throw in an Avocado for free. Avocado on toast makes total sense, why don't they just do it?
Oh I have to pay for that extra? Oh... "brain dead server".
Work required has to be refined and ready. Team have to do the work, write the unit tests, get it reviewed and merged. QE have to accept the update, write their own tests, get that merged and ready for review. PO has to sign off on the delivered work. Promote it into production and check it's ok.
Sounds like at least a full days work to me.
"unable to parse the spaces out of a credit card number"
Add to that - surprise, surprise - DVLA. The V11 has a 16 digit number in groups of 4. The web application accepts a maximum of 16 characters. It doesn't matter whether they can parse spaces or not, there isn't room for them.
I used to have a running log in File Maker it would take an elapsed time and do maths on it calculating pace for eg.
Then system upgrades killed our copy of FM so I tried the DB section of Libre Office which if you put an elapsed time in treats as a time of day, you cannot do maths on it. So I stopped banging my head against did and I do what I tell people not to do. I use the spreadsheet as a DB. There it will treat a time as able to have maths done on it. Go figure.
Yes - it has always bugged me that there's nothing databases or spreadsheets (from my experience, anyway) that deals with time as time - it's always a date part. I deal a lot with elapsed time data, time differences, call lengths, recording lengths, etc. I have to add these up and I don't want 1970-01-12 16:27:23 as the result - I want it in hours, minutes and seconds. So I have to create my own functions and store the results as strings to stop them being reformatted as dates.
"making everything the same colour, having no consistency between different parts of the app, and my personal favourite the pop-under window."
Throw in maximum white space as well. If you only have a few controls on the page it should be possible to see all of them without going full screen and scrolling.
..and if Microsoft (specifically the barely legible Microsoft 365 administration interface that is available in American only, not English)... select a typeface that has really narrow stokes and makes the pathetic text colour vs background problem even worse. Fuck, sometimes I just fire up the Developer Inspector interface and manually override the font with Arial just so I can read the display.
The Evernote forums are full of hissy-fitters like this. Rather than ask for help, they type furious posts about how terrible Evernote has become and that they are leaving for Notepad or some other shite.
I try to respond with "Thanks for letting us know" every time but I can't catch them all.
I had QUITE a few cases where the contact via social meida made things finally work!
Last case, where I am waiting for them to fix it: SK-hynix SSD site. Starts loading, and then error 404. We will se how long. (wait, isn't that a BORK BORK BORK? Maybe being prominently on El Reg News-Site will FINALLY get their asses up to fix this!)
Just tried again with Firefox and Edge on a German Windows 11: Nope, still bugged. Starts loading and then BOOM 404.
W8... video uploaded now.
It also breaks for me...
You're not in the UK or US, are you? Or Japan or Korea... More to the point, your computer's locale isn't English, Japanese, or Korean. (Mine, for example, is Romanian.)
The fault lies in this awesome piece of code, called in an onLoad event, which is redirecting you (us!) to 'unknown/':
function languageRedirect() {
const languageOverride = localStorage.getItem('lang');
const language = languageOverride ? languageOverride : navigator.language.toLocaleLowerCase().slice(0,2);
const langPath = {
'en': '/',
'ja': '/jp',
'ko': '/kr',
}[language];
const path = window.location.pathname;
if (!path.includes(langPath)) {
window.location.replace(langPath+path);
}
}
That's... a spectacular bit of typically incompetent code. The type of code that "works for the developer" because they are American but spectacularly fails for a vast amount of the rest of the world. Just don't get me started on the fact that most American developers are so incompetent they don't know that the Operating System is capable of outputting the date in a format useful for the use and not in North American format only.
To be honest, I'm not sure what astonishes me more - the terrible code, or the fact that someone, somewhere must really be actively ignoring the massive red flags being flown by their server logs and/or analytics.
Makes me feel fairly confident that if I wore a black hat it wouldn't take me very long to find some unpatched vulnerabilities on those servers (they do appear to be using Wordpress, after all...)
> waiting for them to fix it: SK-hynix SSD site. Starts loading, and then error 404.
Inteeresting. VERY slow out here in the woods of Maine, but it does load. There's background images of creepy wizards, platinum_header-1536x864.jpg and gold_header.jpg and gold_header2-1536x864.jpg Not awful-awful large, but several of them, and not much else happens until they cache. The page source is not huge but there's a hellofalotta scripting and layered DIVs.
Probably rendered great in the company meeting.
Another for my 'do not buy' list.
All of the devs should be on a StarLAN-XE ethernet hub (7x12 shared ports) on the other end of an back-to-back pair of ISDN bridges (128k). They should also have 640x400 monitors.
Or we can just* shoot them.
* - google doesn't even have a picute of this switch any more...
This kind of approach to software updates seems to have inspired OnePlus. Over the last month, they have issued something like six OTA updates which are supposed to upgrade various mobiles to their incredibly delayed version of Android 12. The bugs have been many and serious - like not being able to end a call, disconnecting any Bluetooth device after 10 minutes and needing to jump through hoops to get it working again, plus many more entertaining problems.
All clear signs of an approach to testing which consists of "push it out and see how many complaints there are". Thankfully for me in the EU, they have been exclusively inflicting these random builds on the Indian market, which they appear to have unilaterally adopted as their unwilling alpha testers.
Meanwhile, although EU users are not being assaulted with bug ridden builds, they seem to have run out of developers to integrate and push Android security updates to keep our phones safe from exploits targetting the latest known Android security holes.
So to anyone contemplating buying a OnePlus phone, I'd recommend a quick visit to their community forums before you part with your hard earned cash. I sure wish I had done that first!
Thankfully for me in the EU, they have been exclusively inflicting these random builds on the Indian market, which they appear to have unilaterally adopted as their unwilling alpha testers.
Alpha test doesn't just mean earlier than beta. Alpha testing is done by people who have been involved in the development. Beta testing is done by people who haven't i.e. the great unwashed public these days it seems. So the poor sods in India are still beta testers.
Recently aimlessly scrolled through Freeview listings in search of SOMETHING worth watching, and tried Yaas! and it’s sister channel Ketchup Kids. Both channels loaded a Home Screen and then crashed my Sony Android TV (TL;DR - just don’t buy one) and forced a reboot. On to the web chat with Freeview, who said it must be my TV and would I contact Sony. Except Sony last updated this TV 3 years ago. I gently suggested they refer the matter to tech support, perhaps Android 2019 is not over the hill yet.
Freeview are useless - any conversation with them always starts with a 15 minute argument about what transmitter I'm on because their postcode records are wrong for half of West Sussex. Then even if anything turns up purporting to be a fix, it arrives two days after the problem has either sorted itself or been sorted by random button pushing/change in weather conditions/god only knows what it's done
Weather conditions in west Sussex?
I have a hose, I've been watering a garden, I even spilled a little on myself. It's a warm garden today but it's very moist. I could send you a video of me hosing the garden.
I trolled my sister in London with a hosepipe video yesterday. End of the Edinburgh Festival we are searching all English and Welsh visitors for stolen water. West Sussex, northern Sahara.
As I have ended up learning more than them I have not needed to call.
Issues found.
My previous TV was not onCAM compliant from new.
BBC service numbers used to change meaning BBC1 showed Choice and Choice showed 1
All working well on £3 a month IDTV contract then bump.
Anyone need an On Digital CAM?
During all this I ended up on Suttton Coldfield TX.
Now use Freesat as well.
Incompetents list
BBC for service number changes
Pace for diabolical support
Freesat for screwing up Humax PVRs when the most common FS device is a Humax PVR
Ondigital for not getting their CAM specs correct messing up the TV manufacturers.
Good list
Sony for IDTV support
Humax for excellent Freesat kit
My local aerial installer for a good install of both aerial and dish
The attitude I see a lot in software and design can be summed up like this: "I like it and it works for me, therefore you will like it and it will work for you". I call it myopia. The people who make the stuff are special -- just look at their trophy! -- and they know their ideas are great and wonderful. If you don't like it, then the flaw is with you, never them. If it doesn't work for you, then you need to be like them.
Take for example websites. You complain about a website not working. "What browser are you using?" I respond Firefox. Their response: "Just use Google Chrome." In other words, because they use Chrome and they like Chrome they expect you to be just like them. Testing on other browsers. Nah. Just be like me. They use Google Chrome, they have powerful computers, they have ultra-fast internet; ergo, you do too. "Just be like me" is the way they think.
Another example, Windows 7 and earlier would ask you which timezone you live in. Now, everyone lives in the Pacific timezone because Microsoft's home is in the Pacific timezone. Never you mind that the majority of Americans don't even live in the Pacific timezone. "Just be like me" is the way Microsoft thinks. And since they are special -- just look at their two trophies! -- they will suffer no criticism, no matter how legitimate. They cannot understand why you would not like to surrender your personal information to use their product, why you are complaining about drivers installing without your permission, or why you are frustrated that years of muscle memory is being thrown away for a new start menu.
"Just be like me. I have been told all my life that I am special, so it must be true. If there is a flaw, it must be you." I call it myopia. And it is very frustrating.
I think it's worse than "it works OK for me". I don't believe that devs really use the software they create - I mean really, really use all the bells and whistles. No one who's ever had to create and manage documents in Word with corporate fonts, outlining, variable headers and footers, images, contents, footnotes and indexing can ever have had anything to do with developing the bloody programme unless their version of the use-case was to drive users to point of throwing the monitor out of the bloody window.
A major version or two ago, Microsoft intentionally threw out all the central management of styles and document templates instead forcing everything to be locally based or that new documents are instead created in SharePoint and not within the application itself. These were the shared templates and styles directories that got removed in favour of local template and styles only, then checks were added to ensure that creative system managers didn't redirect these local template and style locations to network locations.
In the early 00s I worked for a large company which had a "rebranding". Head office created a set of templates for Word and Powerpoint to make sure that the corporate styles were used properly - they had their own fonts, colours, logos, bullets, etc. all trademarked. This sounded like a great idea, until the templates were sent out. I'd expected proper Word/Powerpoint Template files. Instead, they were just example documents where someone had formatted the Normal text and added logos, bullets etc. to some styles. To create a new Word document you opened the locked sample document, then deleted a load of lorem ispum before starting. Problem was they'd only done the Normal style. So as soon as you created an outline level, header, ToC, or anything that wasn't just a paragraph it defaulted from the expensive company font to Arial. They'd also put the company logo in as a page background, not an image anchored to the page in a header, so it messed up when we went to A4 in the UK or wanted a landscape page. Powerpoint was just as bad.
Turns out they got a quote for creating a proper set of templates, thought it was too expensive so gave it to one of the head office admins to sort out. I bet that millions were spent across the company by people trying to sort it out on an individual basis.
Version 6.
Which was the only browser the SAN management software accepted, which we had to use to, obviously, manage some SANs. Several dozens of them. Which were installed around the time IE6 was EOLed by MS.
Some of them are still around, and in the meantime IE6 has been BANNED by Security.
We just upgraded one of our in-house apps. I say "we". I mean TPTB imposed it on us Some functions don't work in Edge, the "approved" browser, some don't work in Firefox and there's no complete overlap in the Venn diagram. We've been told it fully works in Internet Explorer and to use that instead. Unfortunately, the standard user build doesn't include the now EOLed IE and security policies forbid "unapproved" s/w installs by users, including IE
I've recently been thinking just how ubiquitous this attitude has become.
Having inkjet printer problems? Don't use inkjets. Use laserjets like me. No, I don't need to print. Nor should you!
Trouble with IOS? Use Android. Trouble with Android? Use IOS.
Blah, blah.
I remember the old days was when we tried a bit to help others to achieve their goals. Not continually try to turn them into clones.
I try to tell stories about what works for me. Knowing I'm reasonably excentric. If my story isn't liked, make up your own. Experiments in living, and all of that (J. S. Mill).
Best wishes,
Bob
Don't use inkjets. Use laserjets like me
Problem is, there aren't any inkjets where the manufacturer aren't assholes, like Epson locking the printer after a certain number of cartridges "because the slop pad is full" or companies requiring DRMed ink. And inkjets don't like sitting for a couple months, where my laser just fires up.
If they're really in a situation where the inkjet is either completely incapable of it or cannot do it well enough, then the "switch to laser" reply can be of use, but I'd add a "because" in there. If an inkjet can manage the task, but you don't like them (I've heard many reasons to dislike inkjets, but they do manage to put ink on paper), then the user probably wants to use what they've got, not buy another thing that might also have problems.
I see this as the same problem we see with browsers. Being told to use Chrome when we'd prefer to use Firefox is annoying and often unnecessary, so unless the situation is that Firefox truly can't do what we need, telling us to use Chrome doesn't get us anywhere. Printers have the added difficulty that switching to a laser costs money and takes time, so people may have stronger reasons to want to accomplish their task with what's already available if they can. Adding "You might get better results from a laser printer" at the end is helpful if they do replace it.
The complaints with injets is that if they're used intermittently the ink dries up and they have to replace the cartridge(s) at considerable expense. For that situation a laser is less trouble, it doesn't have that particular problem and in the long term may well be cheaper. IOW the use case may better met by switching and the best advice is to switch and point out that hanging on to the inkjet is the sunk cost fallacy in action.
"May be" because if there isn't enough printer use then the better option might be to take the work to a print shop or a friend with a laser printer or whatever.
The browser situation is easier - just install several. Apart from anything else the trackers don't get the whole story.
This is a perfect example of why I think those comments are not helpful. You've stated some problems with inkkjets. They're probably entirely true (I try to avoid printers because I've never found a printer I didn't have to fight at some point). If you've already got one and it's functioning, using it to print something is not a sunk cost. Buying a laser model or going to a shop would be more expensive, unless the printer is broken or the ink needs replacing (though depending on the model and the user's prediction of how they'll use it later, it might still be cheaper in this situation). The things you've mentioned are very useful to a person who asked "Should I buy an inkjet or a laser to replace this broken inkjet", but not one who is looking for help with their currently functioning inkjet.
And you have the same problem with the browser example: "The browser situation is easier - just install several." I don't want to install Chrome. Being told to do that because you can't be bothered to make a site work in Firefox is not fun, but being told to do so even when Firefox probably would work is even more annoying. I do have several, but I shouldn't need them when a proper site works in all of them.
I worked with a developer who had that mindset. He had huge screen for the time and everything thing he developed he locked down the screen resolution for his screen. Trouble was, only him and a few developers had those screens. And it was critical software for the companyt. Everyone else only saw little pieces of the screen. I should mention, scrollbars were disabled also. His response was to scream at complainers to get a larger screen.
He didn't last long once two VP's had a "chat" with his VP. They pointed out that they had laptops and screens that large with the resolution required weren't available.
My last post on the topic.
Printer: Canon A3 Photoprinter: IP8750
Ink: High quality (on my view) non-OEM.
Cartridges: Canon XL OEM modified to be refillable.
Ancillary: Cartridge resetter. (resets cartridge to reporting full)
Photopaper: A3, plenty of matt, plenty of satin, but a bit low on gloss.
A4, plenty of everything.
Cartridge paper: Plenty of everything. This is what I use for fun, colour, day to day printing.
The Canon XL color cartridges take about 11ml. The non-OEM bottles come with 110ml of ink. I've used about 20ml, so plenty of ink. The set of non-OEM bottles cost a bit more than a set of OEM inks.
The process from raw film to printed output is colour managed.
This reflects what I have fun doing. It is all already paid for.
Oh, and it "just works". For example, put some A4 cartridge paper in the printer, choose cartridge paper as the medium, and it prints the output colour adjusted for that ink and that paper.
Best wishes,
Bob
Sometime n the early 1990s I bought some accounting software. It was unable to reliably produce invoices numbered in consecutive order which is a basic expectation. Creating invoices and getting them paid is the core business function. Everything else is just there to assist this process.
Contacted the company to submit a bug report.
"You need to pay us to fix it, they said".
"I don't want it fixed 'cos I moved to software that actually works. But you might want to fix it for others"
"You still have to pay to submit a bug report".
I gracefully declined. The company was Intuit. Hopefully they have improved their (ex)-customer experience and software.
Reminds me of Greenhills' embedded C/C++ toolchain.
I've only ever had to use it at two places. We'd find compiler bugs and report them for which we were barely even given an acknowledgement after they'd finally agreed that it wasn't that we were 'using the programming language wrong'. If we needed to have them fixed in a timely manner, we were required to pay gouging amounts for a patch.
Greenhills gets used a lot for safety critical because they claim certification to various standards but you have to pay them again to get proof of that.
I've pointed out to management that as gcc is so widely used and open source that is easy to use those facts and a good test suite (which you need anyway regardless of toolchain) to demonstrate its fitness for SC software to the satisfaction of almost all regulatory standards. Even when I show them that I've been able to do that for other clients they still insist on using Greenhills while simultaneously complaining about the engineering budget.
That reminds me doing assessment of a project. Client wanted to split monolith into microservices, because investors said it will look cool for the press release.
Okay.
First thing comes to mind... where are the tests?
"Oh that. Our users help us test the software. If something is wrong, we obviously fix it straight away."
A common practice known as User Test Driven Development.
It never ceases to amaze me that software devs can hold two conflicting ideas: "Well duh, ALL software has bugs, obvs." and "MY code is PERFECT, how DARE you suggest otherwise".
The latter leads them to instantly conclude that problems are caused by the user or whatever else. Support agents are told to stay with that line upon pain of never getting an answer again, so that's what we users get.
But if you describe an "undocumented requirement" in social media then it's often seen as a complaint by the original creators and they will downgrade all conversations that you create to try and solve the problems. Another side of this is that often apps are updated in a way that "works" but illustrates that the new app doesn't understand how the old app was written to work.
> ... it's not a bug, it's an undocumented requirement...
We got one of those years ago after a hospital management software package "upgrade" that changed the password lifetime for every user account to 9999 days. After going through the two-inch thick pile of paper that was the release notes, I found that this change was never mentioned. A call to the vendor went something like "Why would you do that?" to which they responded "Oh... You've instituted an N-day lifetime to comply with HIPAA?" It turns out that they made that password lifetime change at the request of a *single* customer. That it was potentially rolled out to ALL customers lowered my already low opinion of the vendor. And I was left wondering if any other customers had noticed that their user accounts had been mucked with as part of an application upgrade.
So... as another user posted about "kids releasing any old crap" nowadays... hey, it's been going on for years and years.
Here, a user will write a post inviting suggestions to assist with a problem whereupon fellow users try to oblige.
I beg to differ.
The primary function of most forums is for a small group of people to take over and insult anyone who seeks help.
The closest they ever get to 'helping' is to state that the question has already been asked 'lots of times' (note: 'asked' - not necessarily answered), not provide any pointers to the supposed thread(s) (none of which came up when you initially searched), and then get back to hurling insults.
Nine times out of ten, they don't understand what you asked and refer to something else entirely.
"The primary function of most forums is for a small group of people to take over and insult anyone who seeks help."
Not my experience. First they hit you with a ridiculous list of information you must provide before they can possibly deign to help you. Then they insult you.
Hornets appear to get a free pass. Maybe its the similarity of paint job, or just size and pointy stinger, or they are cool and dangerous (from an entomological point of view).
The group does come with a snowflake warning of dead things with pins through them.
You would have thought a hover fly was a better match than and ant, apparently not
Only a fool would confuse a Beefly with a bee.
There is another side.
I occassionally help out on a support group for an open source client tool, trying to pay back a little with time. The tool is primarily used by software developers.
Many requests have clearly made no effort to determine if the same issue has been reported before, where they could join that thread or at least reference it and note how their issue is different. Sometimes the previous queries do have answers, or at least some things to check.
Many requests are a variation of "I tried this thing on my PC and it didn't work". No information about the steps that were performed, the actual outcome, or the expected outcome. They also tend to miss out information about the client software version, the PC (e.g. OS version, which would be relevant to this group), or information about the associated server and network.
Other requests can be about the overall system, and not this particular client. Yes, the users of the client are also users of the generic system but it doesn't seem unreasonable to suggest asking the same question on the support groups for the overall system.
Remember that the user base is primarily software developers!
If someone is about to solicit help from a bunch of volunteers, it seems reasonable that they put in a little effort to prepare the query. I appreciate it's different where a purchase has been made.
The self-righteous expectation that everyone drop everything to help a particular requester can be frustrating. It's sometimes tempting to be blunt, but I'm more likely to simply skip on past to the next query.
"Many requests have clearly made no effort to determine if the same issue has been reported before"
This is not necessarily as straightforward as you might think. The words which user Z uses to describe the problem are not necessarily the same as those used by users A to Y so all those questions are missed. User Y had much the same problem, as did user X...
And FAQs are apt to be, in reality, QWFPETA
Questions We Found Particularly Easy To Answer. or even QWTFAWWRPMWTM
And not helped by the search function on some forum platforms which for a search term like "how to roll back firmware" will find every post with (how) OR (to) OR (roll) OR (back) OR (firmware) - i.e. hundreds of posts with "to" highlighted in blue. If you're really lucky it'll find it in the body as well as the title.
The advanced search will, of course, only search for the exact phrase, so the helpful post "how to rollback firmware" won't be found.
Sounds like the mid-00 IBM web site. Never the same link worked twice. Especially for anything to do with ServeRaid. It was FAR FAR easier to use a site specific google search than IBMs own.
Still far too many web sites like this now days. And a bit of Google-fu and one can search all the web sites. This of course, on the work PC, through a VPN. Because at home it is only DDG.
It can be frustrating to try to report it to the S/W's bug reporting site and find it gets marked as a duplicate of bug xxxxxxx which has been confirmed, then discovered that there are numerous other duplicates and either/or it hasn't been been assigned to a dev/someone has diagnosed the problem,written an tested a patch and it's still not been incorporated.
It sounds like you've been on the Microsoft forums! They work exactly like this, copy-and-paste responses (can't call them answers) from Microsoft Shills which carefully demonstrate that they didn't bother to read the question asked. It seems that the only thing they care about is number of replies so they can stroke their ego.
Ah you mean "reinstall Windows" as the default answer that an equal number of morons then upvote as the solution.
I spent some time trying to figure out why all my sound devices were duffed up and gradually dropping off. Because it was my corporate laptop I was obliged to log a call with our support.
They have no clue but I was confident it was the USB devices (pretty much everything is USB now, even if it is on the systemboard).
Loads of posts on forums with all the same solution, reinstall Windows as that "Fixes" it. Of course is f***ing fixes it, you have reinstalled the OS.
You can delete all the keys from the Registry but it is a royal pain due to permissions and is easier is you use psexec and go in a "System" doing stuff from the CLI. Then I found a neat little utility that did it all for you form nirsoft.
Bingo, one duffed up device was screwing all the sound devices over. Delete them all, reboot and all sorted.
For anyone who does experience these issues, this will safe you days of time if you are not familiar with these sorts of problems:
https://www.nirsoft.net
USBDeview
I have no affiliation, just came across this and it has since fixed similar USB issues for quite a few people.
@CuChulainn
The primary function of most forums is for a small group of people to take over and insult anyone who seeks help.
The closest they ever get to 'helping' is to state that the question has already been asked 'lots of times' (note: 'asked' - not necessarily answered), not provide any pointers to the supposed thread(s) (none of which came up when you initially searched), and then get back to hurling insults.
Nine times out of ten, they don't understand what you asked and refer to something else entirely.
.
My very favourite --- mandatory on Stack-Exchange sites; but wholly traditional in nearly all technical forums --- is a morose moan, dripping with injured stupidity, barking out suspicion:
.
"Why do you want to do that ?" / "I don't understand" / "What are you trying to achieve by this ?"
This.
It may well be that the product doesn't do what the user's trying to do or it does it in a different way*. The first step in problem solving is to work out what the problem is. (The zeroth step is to identify that you have a problem.)
*Example - although I eventually discovered the answer without askingt: The recent addition of mail to Vivaldi clearly shows custom folders on the UI but no obvious way to set them up. It turns out that something which is a click away on TBird can't be done at all in Vivaldi. The "custom folders" just reflect those, if any, which are set up on the server.
Social media is absolutely the worst place to seek assistance with a product
Except for finding hatred or idiots, it's the worst place for quite everything.
All of this, as far as they're concerned, is our fault.
This is sadly a mentality often happening among developers. Ok, they can be right from time to time, but often they are not. But as they take a bug report as a personal attack and threatening their inner core so deeply, their first reaction is to dismiss it (I know, I've been a software developer for a long time ^^). Such a reaction is the other side of their deep implication in their work. Learning them to distantiate, to take a big breath and walk a little before answering is a good start.
The flip side of this is that some of our testers would raise a defect report for a bug in the software if their smoke alarm went off in their kitchen. An innate inability to eliminate the obvious, and determine whether the problem they are facing is actually a bug or not leads to a certain degree of frustration on the part of us devs.
This post has been deleted by its author
If my phone doesn't work and I complain to my phone provider, they don't accuse me of willfully using my phone to call other people, do they?
Please do not give these corps new excuses. Before you know it, the telephone company refuses to accept telephone numbers.
I was working on a shrink-wrapped product. Shipping an update meant literally that - having thousands of CDs made and sent off in the post to many different countries. So, yeah, we put a lot of effort into making sure the damned thing worked. We had regression tests that covered 100% of the code and specific setups to duplicate common user environments.
Kids these days just push any old crap out.
Not only that, they never fix the crap but simply change it for different crap to suit their marketing schedule.
But they have your data and you can't get it back from them, so they know there's nothing you can practically do about it.
Except roll your own, which I'm increasingly convinced would be a better option for large-scale users of common SaaS applications.
So much for shrink wrapped meaning properly tested, in 2009 I licensed a CD mastering product from Sony. On the box it specifically stated the contained software supported a feature I needed. On installation I found that it didn't (the feature was inoperative) - you had to download an upgraded version for the feature to work.
And that is the root cause of so much crap software. It is just too easy to roll out fix after fix, or more usually, fix + new interesting breakage followed by a fix for the fix, inevitably followed by a new bug........
Or maybe I am just too fussy in expecting things to work!
Might I suggest Blind Willie Johnson "If I don't read it, my soul will be lost", apposite blues?
Sure, there are plenty of cowboy developers around, who produce unusable buggy crap, but it's the same in all trades. The TV schedules have been packed for years with "Rogue Builders" type programmes, where viewers are treated to all manner of poor work and can be entertained with tales of plumbers who install baths vertically when they don't fit in the bathroom. This is supposed to educate people into employing trained people with skills and experience.
What we need are some "Rogue Developers" shows, where crap UI design and bug filled atrocities are exhibited to the public for their amusement, and gradually get the message across that it doesn't have to be this way. This might, at best, lead to some increase in quality and lead to more developers actually spending some time studying human-computer interaction and UI design.
Unfortunately, when you do this as a developer, you can then be tripped up by the next level of crappiness : Rogue Management. This is when you're forced to go against any sense of professionalism you have developed and get forced to ship stuff before it's fully written and tested, or when the guy who pays your salary simply insists that all buttons have to be put into the menu ("Don't tell me you can't or shouldn't do that - just do what you''re told"). Then they shamelessly walk into your office after it's released, and the urgent support ticket comes in 5 minutes later asking where all the buttons have gone, and tell you to put all them all back..............
"Rogue Software" just dumping on the product and its vendor would be fine. It doesn't matter whether the fault was the developer or the manglement. It all looks the same to the punter and being shamed might force some vendors to sort themselves out.
> asking where all the buttons have gone, and tell you to put all them all back
At which point we find out if the dev is the sort who will only go by the literal word of the demand and therefore deleted all the old button code, and will now (almost, but not exactly) recreate it from scratch to comply with the new demand. Taking who knows how long to do so.
Or was the sort of dev who left the old button code as it was, just adding in a new boolean flag to disable it; and also used a separate boolean to enable the weird menu behaviour, just in case. He just flicks the boolean, gets the buttons back in minutes.
The big question being: which of those two actually did the better job? In whose eyes?
Was it the one who is being yelled at by Testing because "now they have to check it all works with the flag both both ways"?
Now I am the first to admit that ANY software I have ever supported is far from perfect (if it was perfect I would be unemployed) but I am losing count of the numbers of customers who will swear blind that the cause of the errors is the software that is reporting the errors. The errors have nothing to do with their flaky WAN link, nor their piece-of-salty-wet-string LAN and antique network switches, nor the fact that they are running at 250+% capacity on their VM Hosts. Nooooo, it's always the software.
Sometimes it very well could be the software.
I had a job interview many years ago in front of one of those panels of 5-6 developers where for about 45 minutes every question was "how would you tune the system to <fill-in-the-blank>. After a time I was so tempted to shoot back with "Is all of your software so poorly designed that you rely on OS tuning to make it work acceptably?"
Reminds me of when I had BT broadband years ago (not through choice before anyone starts, my old ISP had folded and the only other option at my exchange at the time was TalkTalk) and I had constant issues with their router. Apparently all of my equipment had DNS issues and it was nothing to do with the edge device that acted as the DNS server, and I was "unhelpful" when I refused to reset said edge device on the grounds that I already had before calling them and doing it again was futile.
Eventually found out the BT HH3 had known issues with the VDSL modem at the time, which BT admitted to, but also refused to provide me with a new HH free of charge and wanted £40 to replace the shoddy hardware they'd given me. Spent nearly £200 on a Draytek router which I still use now and is still supported and receving firmware updates.
My best mate had that same issue with BT when he was with them.
He lives in a small village outside Leeds. At the time, he'd just had broadband installed and it was by BT (which was the only one available).
It was pants!
They must have replaced the microfilters five times to 'fix' the drop outs he kept getting. Back then, the microfilters were the favoured excuse - certainly with BT. And they never did fix the issues.
Some years later he switched to Sky, but Sky uses the same wiring to the exchange, which is by guess who?
I Skype with him several times a week and it is rare we don't get a stuttery connection or broken audio, and common that we have a drop necessitating a reconnect or even a redial. His Sky TV service relies on his broadband, and he often loses connectivity affecting his viewing.
Officially, there is 'no problem'. But my money is on the wet string BT/OpenReach still uses strung between trees to make its connections in many areas.
Doing hardware support for multiple OEMs. One of them has a mandatory procedure to follow which means logging in to their systems and following the diagnosis procedure. You only get to touch their kit if you've already completed their mandatory training. The problem with the online fault diagnosis system is that it's clearly aimed at people who are untrained and have no clue what to do. I have yet to see it come up with a "solution" I could not come up with ten times faster. In theory, on is supposed to carry out each step, one at a time and the click on the result, which then takes you to the next logical step. It even starts with checking the power is on and power cycling. FFS, I KNOW THAT. I DO THAT. IT TAKES A FEW SECONDS. But following their mandatory and monitored online process takes at least 10 minutes just to get that far. The process goes downhill from there :-(
I had to laugh at this. A couple of days ago, occasioned by the unusual fine weather, I got out my drone fleet. (Two of the ilk) I was a bit discomitted to find that the drone controller software crashed a second after loading, for both drones, using different apps. A bit of research revealed that the Android 12 update had broken DJI's software, some months back. There, seemingly had been no action by either party to correct the situation. Dji had already been paid for both devices, and were loth to do anything, since it was not their fault, and who can blame them. A positive outcome for them, was that many people will simply buy the letest offering, which, hopefully, is not affected. In addition, they are bringing out controllers with built in video. So, win,win. Google, on the other hand do what they always do, and ignored the situation, as a take or leave it action. My own solution was to go back to an old Android that is too old to update. But, another solution occured. My usual tablet is fast approaching the time when it has to be permanently plugged into mains power. So, I looked around foe a new tablet running Android 11. Duly, having purchased same, i plugged it in, and before I had any control of the tablet, it dashed off and automatically updated itself to Android 12, and therefore became useless at controlling the drones
"Dji had already been paid for both devices, and were loth to do anything, since it was not their fault, and who can blame them."
I can. Android is a platform and version upgrades are part of the landscape. Vendors who can't or (almost certainly in this case) won't cope with the changes are Just Shit and deserved to be named and shamed at every opportunity. If you want me to pay you money, don't waste your breath offering a product that might stop working at any time and you just wash your hands of the matter.
They want to tell you how unhappy they are with the software
If someone's posting on social meeja about an issue, it usually means they've exhausted all other means of getting a response from the company, and are really angry and desperate. At this point, it's mostly a public statement for others to avoid this POS by all means possible, and they've given up hope.
An unfortunate number of times I've only gotten help after beefing about it on Twibble or whatever.
A lot of companies now employ social media staff. They're probably graduates in media studies (probably be now there are social media studies) and thus enjoy higher status amongst manglement that those unspeakable techie types on user support and development. Unwittingly they've created a mechanism to make them have to actually respond to the customers they've ignored for so long.
Social media is absolutely the worst place to seek assistance with a product.Compare it to a software company's own web-hosted user community forums. Here, a user will write a post inviting suggestions to assist with a problem whereupon fellow users try to oblige. But the same software company's user support groups on social media tend to attract dramatic types.
These are not people seeking help with a software problem. They want to tell you how unhappy they are with the software.
Hmmm.? There are SMARTR Alternative Root Suppliers who would reply .......
Social media is absolutely the worst place to seek assistance with a product.
Compare it to The Register, Biting the hand that feeds IT and Situation Publishing's own web-hosted user community forums. Here, a user will write a post inviting suggestions to assist with a problem whereupon fellow users try to oblige. But similar independent news and views for the tech community company's user support groups on social media will tend to attract all manner of dramatic types.
These are not people seeking help with a software problem or to help with an abiding firmware problem. They want to tell you how unhappy they are with the software and the free licentious sharing of such as may easily be impeccably disruptive wares.
And there is a prime example of use of the facility and utility posted here on El Reg today nearly a full three hours before Dabbsy's contribution to the debate and emerging events ..... with a post inviting peer review and assistance with a Multi-Trillion Dollar Question being tabled and date stamped Fri 12 Aug 2022 // 07:28 UTC
How very coincidental is that. Quite spooky in a stranger things sort of way:-)
And thinking further on the earlier final questions posed there enquiring about the whom and/or the what you would most likely have a bet on having that sort of long-term hypervision and dedicated wherewithall to guarantee at least all of that with life-changing universal deliveries, .... in days of not so long ago yore one would have fully expected it and supported it whenever a royal or imperial prerogative ably exercised and wisely administered.
Are such things greatly changed today? Who then follows whom and for what should ever that not now be the case?
Are there new kids on the block and in the hood doing such things for the greater good?
> I have noticed a growing trend for cloud-based products to enforce updates whether you want them or not. In parallel, there is a growing trend for these mandatory updates to be utterly dreadful.
Ah... sounds just like what smartphone users experience. Our Android-based phones had an 11-to-12 update snuck onto our devices in the middle of the night which made the following morning a rather confusing mess of "WTF! Why does this look/work differently than it did yesterday?" (Of course, there was no warning that this was going to happen so that a proper pre-update backup could be taken.) Sorry, phone carrier... blaming it on the hardware vendor doesn't cut it. I send money to you, the carrier, every month---not the phone manufacturer. The carrier needs to be ensuring that the hardware suppliers are providing good service and updates to their customers. Enough of "the buck stops over there" garbage.
Unless the carrier was responsible for the phone's software, which is sometimes but not often the case, it really isn't their fault. You are paying them frequently, but to provide the connection. I'm assuming your Android devices could still access mobile data, voice, and SMS after the update to Android 12, so they're providing the thing you pay for.
The only way they'd be responsible is if this was a carrier-specific device with their software installed on it. Even if they sold you a device to keep you on a contract, they're not responsible for what runs on it unless it's custom to them. It is, in fact, the manufacturer who you need to complain to or request support from.
> quality software could be written
Yikes, why on earth would you do that? Cheap/barely skilled labor and a short development time with marketing as soon as possible are the keys to better profits. The most blatant bugs can be fixed later on, over a period, while you enjoy your hard-earned bonus.
As an ex- senior support engineer for Applications I would ge quite frustrated at our service desk, and other support engineers My motto I would try to teach was
1 Users are Human
2 Humans Make Mistakes
3. Developers are also Human
From 2. and 3. Derive the forth statement.
I would usually also annoyingly say things when confronted with the "You won't beliieve what X put into that field".. with.. "I don't believe that you allowed X to put that into a field. Do you know what user validation is all for?".
In the world of web UIs, this is really required. no longer can you expect your user to fully undderstand the logic behind the scenes, but you can understand it. So code dfefensively, code knowing what you minimum and maximum values can be, It takes you another 2 or 3 minutes to code but can save hours and days of fault finding. If you are writing an app for someone's PC (rather than Web) then log files can sit on whopping huge hard drives, so use them.
I sometimes feel a bit like the Catweasle character in the '70s TV series who was transported from the 11th century into the 20th. While he was taken by "elec~trickery" he would often complain to his familiar "nothing works, Touchwood."
His technology - magic - largely didn't work in the 20th century and the contemporary technology was incomprehensible.
Even if you hang on to older tech like a Nokia 3G phone that never was able to be updated the Telcos just turn off the network (to free up spectrum for 5G) - as though I would want gigabytes a second streaming into my back pocket.
I can see that running a dead OS (Multics) or some fossilized mainframe OS might have its attractions - not much risk of an unheralded update I should think. Even update announcements by the more glacial Linux distros evoke an automatic "what fresh hell is this?" response.
At least wind-up watches are safe from updates. I was more than surprised that a modern (mumble)shock watch stores the time in UTC and that you have to set/guess the appropriate local timezone.
One might think that would make moving between timezones or onto/off daylight savings easier - it doesn't - its faster to just change the (UTC) time and leave it on LON(don) DST/off.
Even simpler - after killing a couple of mechanical watches by frequently changing the time between timezones - I permanently set it to UTC and just remembered the local offset (its modulo 12 so not too taxing eg AEST+10 = add 12 sub 2 ie just sub 2.)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
