Last week e.g. the Dutch railways announced that from next year on, 100% of their operations will run on electric power from renewable sources, mainly wind, bought from a total of 5 north west European countries
Which is a load of nonsense. They might be paying for that amount of power from those sources but the stuff that actually powers their trains is co-mingled with all the other sources of power including fossil fuels and nuclear. Unless you run dedicated lines from the power supply to the end user there is no way that their trains are only powered by electricity from a specific source.
I've owned & driven Chevy trucks for pretty much my entire life, and never had any sort of issue caused by a manufacturer's defect.
I absolutely guarantee you have had maintenance as a result of a manufacturer's defect if you have truly had Chevy trucks for that long. It may not have been a showstopper problem like this ignition fiasco but it is virtually certain you have had at least one part fail due to a manufacturer's defect. I run a company that makes a lot of auto parts and I've worked for Tier 1 suppliers. They simply don't have the sort of bullet proof quality you might hope for. There isn't a manufacturer on earth that has never shipped a bad batch of parts and there isn't a manufacturer on earth that hasn't unintentionally accepted a bad part and put it into production.
It's not really the engineer's fault for calling a problem a problem, right?
The engineers share some of the blame though the lion's share clearly seems to fall to management. The engineers are to blame to whatever extent they saw a serious problem and did not insist upon appropriate action being taken even in the face of management opposition. Saying "I was just following orders" is an insufficient argument to absolve them of their share of the blame for this fiasco.
I prefer companies that are open about their problems than companies that try to hide problems with "disguised words".
Easy to say when you are not the one at the pointy end of a multi-billion dollar lawsuit. Lots of people have plenty of courage in a semi-anonymous internet post. While I agree with you in principle the way the laws are written it isn't nearly as simple as you or I think it should be. As much as I'd like to see engineers speaking freely about problems, the consequences of doing so can be catastrophic when they don't know what they are doing. And I don't know too many engineers who are up to date on their product liability law.
Fact is that NO lawyer worth his retainer would agree with you. The number of ways in which employees can get a company in serious financial trouble through even the most honest attempts to solve problems is HUGE. Employees can agree to contracts, "admit" to wrongdoing (even when there wasn't any), etc. There are VERY good reasons why companies tend to only let a few, carefully selected people who know what they are doing speak for the company. I've worked as an engineer at a large auto company and I had to get special permission to give a technical talk just due to the potential liability and trade secret issues involved.
You know, if you buy hardware and a support contract, you want to download the latest (and/or greatest) image for your device and reflash it yourself anyway.
If the hardware you cannot flash has been compromised you can flash whatever you want and it probably won't matter.
That's a very shortsighted view and brought to you by the same people who tell you that inflation is a good thing as they help themselves to the wealth you earned.
If you think it is shortsighted, prove where my logic is wrong. Show how weakening currencies do not actually aid exports and strengthening currencies do not aid imports. When you do that, publish some peer reviewed papers and collect your Nobel Prize. (because if you actually can prove that you will win one)
Furthermore, nobody says "inflation is a good thing". Ideally we would want neither inflation nor deflation but that is impossible to ensure for a lot of reasons including because human populations do not remain constant even if you hold the money supply fixed. We know from experience that a lot of inflation or deflation is very bad. We also know from experience that the effects of a little inflation is not as bad as the effects of a little deflation. So we try our best to ensure there is a modest amount of inflation, somewhere between 0%-5%. If you think I'm wrong, prove it, publish it, and collect your second Nobel Prize.
The phone also allows new OS software only if it is signed by Apple. That should be equally possible on an Cisco router.
If Cisco can monitor your gear, so can the NSA. You are presuming that Cisco actually is not in cahoots with the NSA. While it is certainly possible Cisco is not working with the NSA, a foreign buyer cannot assume that is true because they have no way to confirm.
Problem is that there is pretty much no possible way Cisco can put the toothpaste back in the tube. They have no simple way to prove to potential customers that their gear hasn't been hacked or compromised in some way. The actions (real or perceived) of the NSA have basically screwed a number of US companies in overseas markets where security is any sort of a concern.
Basically even the perception that the NSA may have compromised the equipment is enough to keep people from buying Cisco. Of course then the question becomes who do you trust? The Chinese make a lot of gear but they are probably trusted even less than the Americans if anything. Unless the gear is manufactured domestically under supervision it's unclear how you ensure that no one has introduced undesirable code/hardware.
Maybe it really isn't shit today and the US currency has actually devalued this much?
I think you may be confused about what currency devaluation means relative to other currencies. The mere fact that a dollar buys fewer Euros than it used to is only bad if you are trying import goods. If you are exporting goods (and we do a lot of that) then it means your products are more competitive. The Chinese have intentionally kept their currency relatively "weak" compared to the dollar in large part because it makes their exports less expensive compared with the competition. Japanese car companies have been doing fairly well lately in large part because the Yen has depreciated significantly against other major currencies.
A "weak" currency is only a bad thing if you are trying to reduce exports and increase imports. If you want to reduce the trade gap with China then you WANT a somewhat weaker dollar and a stronger yuan. If you want US companies to be able to sell their products overseas then you WANT a weaker dollar.
the prices paid for "strategic acquisitions" are way inflated
I would certainly argue that some of the acquisitions I've seen lately (looking at you Facebook) are WAY too expensive. Facebook buying WhatsApp for $19 billion makes zero economic sense that I can discern. I don't know enough about Twitch to properly evaluate this acquisition but the valuation to a casual observation seems a bit rich for a website that hasn't been in existence even 3 years yet. Google certainly has the money and their track record isn't bad but if I were a shareholder I'd seriously wonder how careful Google was being with cash.
I'm slightly confused.... is Youtube buying Twitch or is Google's wholly owned subsidiary (Youtube) buying Twitch?
It's a distinction without a difference. YouTube is a wholly owned subsidiary of Google. It would be equally accurate to say YouTube is buying Twitch, Google's subsidiary YouTube is buying Twitch, or Google is buying Twitch. In the end equation they all mean the same thing for all practical purposes. There are some subtle accounting ramifications regarding whether YouTube or Google actually is the the buying entity but nothing you or I will care about in the slightest.
Why would regulators care at all about this deal? Twitch isn't a public company.
Whether a company is traded publicly or not is irrelevant to anti-trust concerns. The only thing that being a publicly traded company means is that the stock is traded on an exchange. That's all. Many large companies are not publicly traded and anti-trust regulators are concerned with whether the merger will adversely affect consumers and competition in the market. Whether the stock is traded on a stock exchange is completely unimportant to the analysis.
The point is that the claim of consensus is meaningless.
Consensus is not meaningless if it is a second order effect of the evidence for a particular theory. For example general relativity has been tested heavily and so far every piece of evidence shows that it is a very accurate model. In the face of this evidence a consensus has developed that general relativity is "correct". Consensus by itself means nothing unless it follows from the natural and proper progression of scientific inquiry.
The problem is that some people misunderstand that consensus means an agreement of opinion similar to political opinions. It means nothing of the sort in scientific circles. It means that the available evidence overwhelmingly supports a hypothesis to the degree that further disagreement would be foolish without extraordinary evidence.
I've heard this story before. About a million times on TDWTF.
I'm pretty sure we all have. I'm well aware I'm not the first person to criticize database front end tools. Nevertheless the problem remains... Sometimes simple is really hard to do.
I would try to avoid states that pass laws like this. Clearly this is a Federal matter.
And Congress has been shown themselves to be right on top of this issue too... [/sarcasm]
Congress has shown zero appetite for dealing with this matter. Until they can actually be bothered to do something in the interest of the country I'm fine with states taking up the slack where they can.
"Price per gigabyte" is maybe a better way to put it. It's not relating to the cost of provoding the service, it's just how they charge out you accessing the service.
The cost of actually transmitting the data to the people you pay your bill to is a tiny percentage of that bill. 80-95% of your bill is for things other than the actual cost of transmission. The important thing to understand is the concept of tax incidence - namely, at the end of the day who is really paying for this stuff. For a stupidly simple example, if Level 3 raises their rates on Comcast, Comcast is going to pass that cost on to you and me. It doesn't really matter what Level 3 does because that is effectively a transfer payment. It's like when you buy a car, you really are buying a product that several hundred companies had a hand in making but how they divide up your car payment among themselves really isn't the important thing.
The backhaul providers only charge on a per-GB basis because their actual costs are mostly fixed and charging per-GB is a back of the envelope way of correlating resources required with usage. Think of it kind of like using a gas tax to pay for road maintenance. You can't assign the costs with much accuracy so you use a proxy for usage. It's pretty crude but it works well enough. It's the sort of problem that keeps cost accountants employed.:-)
Do you enjoy wasting time needlessly with inefficient badly designed tools?
Think of it as constructive laziness in the sense of the three virtues of a good programmer. I have a lot to do and anything that helps me spend less time screwing around with details that don't matter for the task at had is a Good Thing.
I've developed databases in both and when developing small custom databases it most assuredly is NOT the same aside from the fact that both are databases.
If you want people who don't know what the fuck tables and indexes are to create tables and indexes, you've fucked up.
You misunderstand. What I want is a MUCH better interface for making tables and indexes and linking them together in basic ways, particularly for small ad-hoc databases. For example my company has a small database we use to track tooling - there are no off the shelf bits of software that can do what we need. We prototyped it in a spreadsheet to figure out the basic structure and now it's implemented using Access which is fine. I could have used PostgreSQL (I've done it before) but that would have been overkill. It's not a complicated database - a few tables and some simple queries plus 3 forms. But the available tools for prototyping and then implementing it are harder than necessary. Hard in the sense of unnecessarily clumsy. Hard in the sense of worrying about too many fussy details that simply don't matter for such a simple project. Hard in the sense that making simple adjustments to such a simple tool requires more technical chops than is economically sane.
Access is "harder to use than it should be" because you're looking at a thing you don't understand and trying to put it to use.
You are making the mistake of confusing criticism with ignorance. I don't criticize because I don't understand I criticize because I am the one who actually has to implement this stuff in the real world in places where IT staffs are simply not available and cannot be afforded. I understand Access just fine and use it with more regularity than I care to think about. I've also had to implement numerous small custom ad-hoc databases in all sorts of companies for the last 15 years. I've spent a good part of my career basically as a work efficiency consultant and I run a small manufacturing company. I'm the person who has to actually implement this stuff in companies that don't have full time IT staff and can't hire a DBA. And what I'm telling you is that the back end of the tools are fine. The development front end for pretty much every single database product SUCKS for making basic tools.
There are lots of situations where this is simply not an option. You should be able to do simple real world database work without requiring the assistance of a DBA. The inability to do this speaks to the poor quality of the interfaces available for database creation.
The requirement to allow non-technical people to perform technical tasks without the knowledge and experience is a critical modern mistake.
I disagree. I think it is a deficiency in the quality of the tools that the technical people have made thus far. Hell I *am* a technical person and I would dearly love for a database that made it as easy to create a simple database as it is to create a spreadsheet. I have needs for small databases all the time but even MySQL is WAY overkill. Access the closes thing to usable but even it is hard to use than it should be. While I'm perfectly capable of making it work it takes FAR longer than it should.
Spreadsheets are not databases, and should never be used as such.
Spreadsheets will stop being used as databases when someone invents a database that is as easy to use as a spreadsheet. Nobody has done this yet. I've never seen a database that was even close to as easy to use as a spreadsheet. Even Access is still MUCH harder to use than a spreadsheet. MySQL & PostgreSQL are severe overkill for a lot of projects. Openoffice Base is useful as an interface between real databases and spreadsheets (I use it for that) but not much more.
Spreadsheets can be simple databases (that's basically what a table is after all) and actually aren't a bad way to prototype a (very) simple database. I do this all the time. I would KILL for a tool that allowed me to easily convert a spreadsheet table into a real database table and make creating forms much easier than it is now.
So if people went for the deal where they would just put up with a possible 1-2 day power outage per year they would get electricity at a huge discount (50% or more IIRC). However those 1 or 2 days are the hottest days in the year, and with no air con or fridge... Most went off the plan after that deciding that paying 2x more for that "uptime" guarantee was worth it.
I'd consider doing that if I had a whole house generator attached to my gas line. Might actually pay for the generator over time if the usage really was only 1-2 days per year.
One could make a case that its not a correct way of billing or charging
It's an ok way to bill as long as you don't confuse the revenue model with the cost model. Most of the costs of the telecom/cable providers are fixed costs including hardware, maintenance, sales, overhead, electricity, etc. They will have to pay these each month even if they don't send a single byte of data. These fixed costs have to be allocated to each customer in some fashion and one way to do it is to charge a pro-rata per-byte rate that is high enough to cover the variable + fixed costs. It doesn't do a good job of allocating costs streams to revenue streams but it's arguably not worse than many other revenue models.
It's actually a nearly impossible exercise to allocate precisely the amount of fixed costs attributable to each customer. How much electricity at headquarters should be allocated to each customer for instance? Anyone who tells you the answer is straightforward doesn't understand cost accounting.
Everyone north of Tier 1 providers pay per gigabyte of data they transfer over Tier 1 backbones.
The cost per gigabyte of data is a relatively small percent of the cost incurred by companies like Comcast. Comcast's gross margins are somewhere around 70% and the cost per gigabyte would properly be accounted for under Cost of Goods Sold (also called Cost of Revenue - second line on the income statement under Revenue). Comcast's net margins are around 10% which means that the cost of moving data cannot account for more than around 30% of their expenses even if every penny of COGS was used to pay for data - which it definitely is not. In reality the real number as a percent of their total expense is probably somewhere around 10-15% at most. It's not trivial but it is a small percentage of what they actually pay out each period.
Paying per gigabyte is how the internet actually works;
I assure you that the cost per gigabyte is only a relatively minor portion of the costs involved and realistically it's not the biggest one. Don't confuse revenue models with cost models. Most of the actual cost of the internet (far in excess of 50%) is in hardware, maintenance, electricity, sales and overhead. All of these are fixed costs, unaffected by your data usage. Don't take my word for it, look on their financial statements yourself.
The vast majority of ISPs in this country do not offer any (or very little) TV service at all.
You must not live in the US because Comcast, Charter, Time/Warner, Cox, AT&T and Verizon all offer TV service and they are by a huge margin the largest ISPs. The top 10 ISPs account for somewhere around 75% of the market. Comcast alone has somewhere north of 20% of the market by themselves.
I've recently switched to Acanac where I'm paying less than $50 for the same speeds with no cap. Hopefully US customers will be able to find smaller/independant ISPs that offer something similar...
Not a chance. I have precisely 3 sets of wires coming into my house. Cable company, phone company and the power company which doesn't provide data services. I don't need land line phone services and the only internet provider capable of better than slow DSL speeds in my area is Comcast. Any "third party" would have to come across one of those sets of wires so I'm stuck dealing with the same companies at some level regardless of who actually sends me the bill.
Fortunately Comcast hasn't been evil to me (knock on wood). Expensive for their TV but I get 100Mbit internet for less than $100US per month. Their TV is outrageously expensive for what you get but I don't watch a huge amount of TV so I don't spend much on that.
Slashdot Mirror
Last week e.g. the Dutch railways announced that from next year on, 100% of their operations will run on electric power from renewable sources, mainly wind, bought from a total of 5 north west European countries
Which is a load of nonsense. They might be paying for that amount of power from those sources but the stuff that actually powers their trains is co-mingled with all the other sources of power including fossil fuels and nuclear. Unless you run dedicated lines from the power supply to the end user there is no way that their trains are only powered by electricity from a specific source.
I've owned & driven Chevy trucks for pretty much my entire life, and never had any sort of issue caused by a manufacturer's defect.
I absolutely guarantee you have had maintenance as a result of a manufacturer's defect if you have truly had Chevy trucks for that long. It may not have been a showstopper problem like this ignition fiasco but it is virtually certain you have had at least one part fail due to a manufacturer's defect. I run a company that makes a lot of auto parts and I've worked for Tier 1 suppliers. They simply don't have the sort of bullet proof quality you might hope for. There isn't a manufacturer on earth that has never shipped a bad batch of parts and there isn't a manufacturer on earth that hasn't unintentionally accepted a bad part and put it into production.
It's not really the engineer's fault for calling a problem a problem, right?
The engineers share some of the blame though the lion's share clearly seems to fall to management. The engineers are to blame to whatever extent they saw a serious problem and did not insist upon appropriate action being taken even in the face of management opposition. Saying "I was just following orders" is an insufficient argument to absolve them of their share of the blame for this fiasco.
Please... In my humble opinion every single lawyer needs to be shot on sight .
What a mature attitude you have there. [/sarcasm] Better hope you never find yourself in need of legal assistance.
I prefer companies that are open about their problems than companies that try to hide problems with "disguised words".
Easy to say when you are not the one at the pointy end of a multi-billion dollar lawsuit. Lots of people have plenty of courage in a semi-anonymous internet post. While I agree with you in principle the way the laws are written it isn't nearly as simple as you or I think it should be. As much as I'd like to see engineers speaking freely about problems, the consequences of doing so can be catastrophic when they don't know what they are doing. And I don't know too many engineers who are up to date on their product liability law.
Fact is that NO lawyer worth his retainer would agree with you. The number of ways in which employees can get a company in serious financial trouble through even the most honest attempts to solve problems is HUGE. Employees can agree to contracts, "admit" to wrongdoing (even when there wasn't any), etc. There are VERY good reasons why companies tend to only let a few, carefully selected people who know what they are doing speak for the company. I've worked as an engineer at a large auto company and I had to get special permission to give a technical talk just due to the potential liability and trade secret issues involved.
You know, if you buy hardware and a support contract, you want to download the latest (and/or greatest) image for your device and reflash it yourself anyway.
If the hardware you cannot flash has been compromised you can flash whatever you want and it probably won't matter.
That's a very shortsighted view and brought to you by the same people who tell you that inflation is a good thing as they help themselves to the wealth you earned.
If you think it is shortsighted, prove where my logic is wrong. Show how weakening currencies do not actually aid exports and strengthening currencies do not aid imports. When you do that, publish some peer reviewed papers and collect your Nobel Prize. (because if you actually can prove that you will win one)
Furthermore, nobody says "inflation is a good thing". Ideally we would want neither inflation nor deflation but that is impossible to ensure for a lot of reasons including because human populations do not remain constant even if you hold the money supply fixed. We know from experience that a lot of inflation or deflation is very bad. We also know from experience that the effects of a little inflation is not as bad as the effects of a little deflation. So we try our best to ensure there is a modest amount of inflation, somewhere between 0%-5%. If you think I'm wrong, prove it, publish it, and collect your second Nobel Prize.
The phone also allows new OS software only if it is signed by Apple. That should be equally possible on an Cisco router.
If Cisco can monitor your gear, so can the NSA. You are presuming that Cisco actually is not in cahoots with the NSA. While it is certainly possible Cisco is not working with the NSA, a foreign buyer cannot assume that is true because they have no way to confirm.
Problem is that there is pretty much no possible way Cisco can put the toothpaste back in the tube. They have no simple way to prove to potential customers that their gear hasn't been hacked or compromised in some way. The actions (real or perceived) of the NSA have basically screwed a number of US companies in overseas markets where security is any sort of a concern.
Basically even the perception that the NSA may have compromised the equipment is enough to keep people from buying Cisco. Of course then the question becomes who do you trust? The Chinese make a lot of gear but they are probably trusted even less than the Americans if anything. Unless the gear is manufactured domestically under supervision it's unclear how you ensure that no one has introduced undesirable code/hardware.
Maybe it really isn't shit today and the US currency has actually devalued this much?
I think you may be confused about what currency devaluation means relative to other currencies. The mere fact that a dollar buys fewer Euros than it used to is only bad if you are trying import goods. If you are exporting goods (and we do a lot of that) then it means your products are more competitive. The Chinese have intentionally kept their currency relatively "weak" compared to the dollar in large part because it makes their exports less expensive compared with the competition. Japanese car companies have been doing fairly well lately in large part because the Yen has depreciated significantly against other major currencies.
A "weak" currency is only a bad thing if you are trying to reduce exports and increase imports. If you want to reduce the trade gap with China then you WANT a somewhat weaker dollar and a stronger yuan. If you want US companies to be able to sell their products overseas then you WANT a weaker dollar.
the prices paid for "strategic acquisitions" are way inflated
I would certainly argue that some of the acquisitions I've seen lately (looking at you Facebook) are WAY too expensive. Facebook buying WhatsApp for $19 billion makes zero economic sense that I can discern. I don't know enough about Twitch to properly evaluate this acquisition but the valuation to a casual observation seems a bit rich for a website that hasn't been in existence even 3 years yet. Google certainly has the money and their track record isn't bad but if I were a shareholder I'd seriously wonder how careful Google was being with cash.
I'm slightly confused.... is Youtube buying Twitch or is Google's wholly owned subsidiary (Youtube) buying Twitch?
It's a distinction without a difference. YouTube is a wholly owned subsidiary of Google. It would be equally accurate to say YouTube is buying Twitch, Google's subsidiary YouTube is buying Twitch, or Google is buying Twitch. In the end equation they all mean the same thing for all practical purposes. There are some subtle accounting ramifications regarding whether YouTube or Google actually is the the buying entity but nothing you or I will care about in the slightest.
Why would regulators care at all about this deal? Twitch isn't a public company.
Whether a company is traded publicly or not is irrelevant to anti-trust concerns. The only thing that being a publicly traded company means is that the stock is traded on an exchange. That's all. Many large companies are not publicly traded and anti-trust regulators are concerned with whether the merger will adversely affect consumers and competition in the market. Whether the stock is traded on a stock exchange is completely unimportant to the analysis.
The point is that the claim of consensus is meaningless.
Consensus is not meaningless if it is a second order effect of the evidence for a particular theory. For example general relativity has been tested heavily and so far every piece of evidence shows that it is a very accurate model. In the face of this evidence a consensus has developed that general relativity is "correct". Consensus by itself means nothing unless it follows from the natural and proper progression of scientific inquiry.
The problem is that some people misunderstand that consensus means an agreement of opinion similar to political opinions. It means nothing of the sort in scientific circles. It means that the available evidence overwhelmingly supports a hypothesis to the degree that further disagreement would be foolish without extraordinary evidence.
I've heard this story before. About a million times on TDWTF.
I'm pretty sure we all have. I'm well aware I'm not the first person to criticize database front end tools. Nevertheless the problem remains... Sometimes simple is really hard to do.
I would try to avoid states that pass laws like this. Clearly this is a Federal matter.
And Congress has been shown themselves to be right on top of this issue too... [/sarcasm]
Congress has shown zero appetite for dealing with this matter. Until they can actually be bothered to do something in the interest of the country I'm fine with states taking up the slack where they can.
"Price per gigabyte" is maybe a better way to put it. It's not relating to the cost of provoding the service, it's just how they charge out you accessing the service.
The cost of actually transmitting the data to the people you pay your bill to is a tiny percentage of that bill. 80-95% of your bill is for things other than the actual cost of transmission. The important thing to understand is the concept of tax incidence - namely, at the end of the day who is really paying for this stuff. For a stupidly simple example, if Level 3 raises their rates on Comcast, Comcast is going to pass that cost on to you and me. It doesn't really matter what Level 3 does because that is effectively a transfer payment. It's like when you buy a car, you really are buying a product that several hundred companies had a hand in making but how they divide up your car payment among themselves really isn't the important thing.
The backhaul providers only charge on a per-GB basis because their actual costs are mostly fixed and charging per-GB is a back of the envelope way of correlating resources required with usage. Think of it kind of like using a gas tax to pay for road maintenance. You can't assign the costs with much accuracy so you use a proxy for usage. It's pretty crude but it works well enough. It's the sort of problem that keeps cost accountants employed. :-)
Are you sure you are "a technical person"?
Do you enjoy wasting time needlessly with inefficient badly designed tools?
Think of it as constructive laziness in the sense of the three virtues of a good programmer. I have a lot to do and anything that helps me spend less time screwing around with details that don't matter for the task at had is a Good Thing.
Using PostgreSQL is the same as using Access
I've developed databases in both and when developing small custom databases it most assuredly is NOT the same aside from the fact that both are databases.
If you want people who don't know what the fuck tables and indexes are to create tables and indexes, you've fucked up.
You misunderstand. What I want is a MUCH better interface for making tables and indexes and linking them together in basic ways, particularly for small ad-hoc databases. For example my company has a small database we use to track tooling - there are no off the shelf bits of software that can do what we need. We prototyped it in a spreadsheet to figure out the basic structure and now it's implemented using Access which is fine. I could have used PostgreSQL (I've done it before) but that would have been overkill. It's not a complicated database - a few tables and some simple queries plus 3 forms. But the available tools for prototyping and then implementing it are harder than necessary. Hard in the sense of unnecessarily clumsy. Hard in the sense of worrying about too many fussy details that simply don't matter for such a simple project. Hard in the sense that making simple adjustments to such a simple tool requires more technical chops than is economically sane.
Access is "harder to use than it should be" because you're looking at a thing you don't understand and trying to put it to use.
You are making the mistake of confusing criticism with ignorance. I don't criticize because I don't understand I criticize because I am the one who actually has to implement this stuff in the real world in places where IT staffs are simply not available and cannot be afforded. I understand Access just fine and use it with more regularity than I care to think about. I've also had to implement numerous small custom ad-hoc databases in all sorts of companies for the last 15 years. I've spent a good part of my career basically as a work efficiency consultant and I run a small manufacturing company. I'm the person who has to actually implement this stuff in companies that don't have full time IT staff and can't hire a DBA. And what I'm telling you is that the back end of the tools are fine. The development front end for pretty much every single database product SUCKS for making basic tools.
Just use PostgreSQL from the outset
There are lots of situations where this is simply not an option. You should be able to do simple real world database work without requiring the assistance of a DBA. The inability to do this speaks to the poor quality of the interfaces available for database creation.
The requirement to allow non-technical people to perform technical tasks without the knowledge and experience is a critical modern mistake.
I disagree. I think it is a deficiency in the quality of the tools that the technical people have made thus far. Hell I *am* a technical person and I would dearly love for a database that made it as easy to create a simple database as it is to create a spreadsheet. I have needs for small databases all the time but even MySQL is WAY overkill. Access the closes thing to usable but even it is hard to use than it should be. While I'm perfectly capable of making it work it takes FAR longer than it should.
Spreadsheets are not databases, and should never be used as such.
Spreadsheets will stop being used as databases when someone invents a database that is as easy to use as a spreadsheet. Nobody has done this yet. I've never seen a database that was even close to as easy to use as a spreadsheet. Even Access is still MUCH harder to use than a spreadsheet. MySQL & PostgreSQL are severe overkill for a lot of projects. Openoffice Base is useful as an interface between real databases and spreadsheets (I use it for that) but not much more.
Spreadsheets can be simple databases (that's basically what a table is after all) and actually aren't a bad way to prototype a (very) simple database. I do this all the time. I would KILL for a tool that allowed me to easily convert a spreadsheet table into a real database table and make creating forms much easier than it is now.
So if people went for the deal where they would just put up with a possible 1-2 day power outage per year they would get electricity at a huge discount (50% or more IIRC). However those 1 or 2 days are the hottest days in the year, and with no air con or fridge... Most went off the plan after that deciding that paying 2x more for that "uptime" guarantee was worth it.
I'd consider doing that if I had a whole house generator attached to my gas line. Might actually pay for the generator over time if the usage really was only 1-2 days per year.
One could make a case that its not a correct way of billing or charging
It's an ok way to bill as long as you don't confuse the revenue model with the cost model. Most of the costs of the telecom/cable providers are fixed costs including hardware, maintenance, sales, overhead, electricity, etc. They will have to pay these each month even if they don't send a single byte of data. These fixed costs have to be allocated to each customer in some fashion and one way to do it is to charge a pro-rata per-byte rate that is high enough to cover the variable + fixed costs. It doesn't do a good job of allocating costs streams to revenue streams but it's arguably not worse than many other revenue models.
It's actually a nearly impossible exercise to allocate precisely the amount of fixed costs attributable to each customer. How much electricity at headquarters should be allocated to each customer for instance? Anyone who tells you the answer is straightforward doesn't understand cost accounting.
Disclosure: I'm a certified cost accountant.
Everyone north of Tier 1 providers pay per gigabyte of data they transfer over Tier 1 backbones.
The cost per gigabyte of data is a relatively small percent of the cost incurred by companies like Comcast. Comcast's gross margins are somewhere around 70% and the cost per gigabyte would properly be accounted for under Cost of Goods Sold (also called Cost of Revenue - second line on the income statement under Revenue). Comcast's net margins are around 10% which means that the cost of moving data cannot account for more than around 30% of their expenses even if every penny of COGS was used to pay for data - which it definitely is not. In reality the real number as a percent of their total expense is probably somewhere around 10-15% at most. It's not trivial but it is a small percentage of what they actually pay out each period.
Paying per gigabyte is how the internet actually works;
I assure you that the cost per gigabyte is only a relatively minor portion of the costs involved and realistically it's not the biggest one. Don't confuse revenue models with cost models. Most of the actual cost of the internet (far in excess of 50%) is in hardware, maintenance, electricity, sales and overhead. All of these are fixed costs, unaffected by your data usage. Don't take my word for it, look on their financial statements yourself.
The vast majority of ISPs in this country do not offer any (or very little) TV service at all.
You must not live in the US because Comcast, Charter, Time/Warner, Cox, AT&T and Verizon all offer TV service and they are by a huge margin the largest ISPs. The top 10 ISPs account for somewhere around 75% of the market. Comcast alone has somewhere north of 20% of the market by themselves.
I've recently switched to Acanac where I'm paying less than $50 for the same speeds with no cap. Hopefully US customers will be able to find smaller/independant ISPs that offer something similar...
Not a chance. I have precisely 3 sets of wires coming into my house. Cable company, phone company and the power company which doesn't provide data services. I don't need land line phone services and the only internet provider capable of better than slow DSL speeds in my area is Comcast. Any "third party" would have to come across one of those sets of wires so I'm stuck dealing with the same companies at some level regardless of who actually sends me the bill.
Fortunately Comcast hasn't been evil to me (knock on wood). Expensive for their TV but I get 100Mbit internet for less than $100US per month. Their TV is outrageously expensive for what you get but I don't watch a huge amount of TV so I don't spend much on that.