The negative attitude is what stops ideas from flourishing.
You have no complete files on your computer, just random bits and blocks. At some point, the compression techniques will improve to ensure that data can be reconstructed from these disparate pieces.
Bandwidth is transient, you use your bandwidth, and it's just a passive background task that's always going onwards, and as more units come online into this world, the less bandwidth each unit is required. The files and file systems are designed to handle losses, machines and boxes dying. The blocks are spread across multiple devices and are always kept redundant, so that the chances of data becoming lost approaches zero. If the primary source of your target block is gone, route to another storage loaction of that particular block.
It's is not about large companies supplying all the space (albeit, some will) it's about -every- device offering up space, cpu, hardward, and bandwidth to ensure that the flow of information can never be stopped, becuase it's ubiquitous.
It's for the the full mobility of the data everywhere. Beyond a single tier provider like drop-box, but spread across every infrastructure storage. And as advances in storgage increase, those would be deployed and handled the increased load.
The vision is to ensure that day-to-day life goes on, but everyone still needs a hard-drive for local offline storage, boot-up, caching, and the like. Storage media will not be disappearing so long as ubiquitious internet access world-wide does not exist. Thus, there will be a market, companies will still have infrastructure. It's about having every device in the world suppling redundance storage space for multiple copies of data.
Combine to form a single displayable character. (Sri) (kinda a fancy item; but different from without the ZWJ which would display two graphemes. (S', and RII)
And Lituanian: "However, not all abstract characters are encoded as a single Unicode character, and some abstract characters may be represented in Unicode by a sequence of two or more characters. For example, a Latin small letter "i" with an ogonek, a dot above, and an acute accent, which is required in Lithuanian, is represented by the character sequence U+012F, U+0307, U+0301."
And there are many other cases where there is no single code-point to represent a single grapheme. So for string truncation and line-splitting, (and anything dealing with arabic or indic scripts), you need to never crop in the middle of a codepoint-sequence that defined a single grapheme; or else the visual display is incorrect, or bakamoji (jibbrish).
Characters in Thai are rendered in display-oredr, and not logical order. so, for example ( mina would be imna) and requires reordering for sorting.
Characters in many Indic languages are still all syllable based. So, consonants and vowels are encoded separately, and fully interact as a logical graphical character.
Combine to form a single displayable character. (Sri)
If you omit the Zero-Width-Joiner, then it displays as two characters, "Sa'" and "Ri." So, the rendering and display are dependant on the entire grapheme, which is the normal unit of display and truncation. Otherwise one will be cropping portions of a character on display; and rendering either jibbrish/bakamoji, or unrelated characters/syllables because.
Malay: 0d15 0d4d 0d38 0d3e KA VIRAMA SA AA
One displayable character. If you display code-point by code point, the grapheme displayed would changes 4 times. KA K' KSA KSAA
"However, not all abstract characters are encoded as a single Unicode character, and some abstract characters may be represented in Unicode by a sequence of two or more characters. For example, a Latin small letter "i" with an ogonek, a dot above, and an acute accent, which is required in Lithuanian, is represented by the character sequence U+012F, U+0307, U+0301."
THe arguments above are missing the point of this development. It's fear that's the root of all evils, and prevention of advancement. And the fear, however irrational/illogical on your personal scale is the only obstacle of advancement for all.
This isn't just about putting others' files on your computer, in broken encrypted pieces. It isn't about the legal ramifications of having random unreadable bits on your hard-drive shard. What this is, is the future of truly unlimited storage for everyone. By creating a P2P storage solution, it's creating an "Internet" of storage space that everyone in the world can use. It can render the local hard-drive solution solely a "cache" of files, but all the files, all the items you access will live on across the network.
If it's done correctly, It will allow one to lose your hard-drive completely, and have all your files instantly available. Available from any computer interface anywherer in the world at any time. And, depending on your decryption keys, or more specficially, your custom data-access identifier, you can have multiple file-stores, that are independant, and not related in any way. Or even co-mingling.
This has the prospect of leading the future into a truly data-everywhere situation. The only item that needs to be resolved, is how to make this information publically available after some time. History is being lost by the Encryption, and the loss of private journals, of private note-writings, and such. And over time, it is those items that need to be protected and spread across the world to give insight into who you were, and into your thoughts, dreams, and different views on the events of your lifetime.
But, that can be handled after. By doing this, we can pretty much guarantee that information can never be lost again, (which is different from ever being exposed.) Which is a good thing.
I got a ticket doing 65 in the left-most lane; Reason: Car coming up behind me had to break. It's an enforcable law in most states, that left lane is for active passing only. More so in Utah, apparently.
They are using ISO Year for the Date header, for some reason. (the last 3 years wouldn't have been affected) As Mon Dec 29, 2014, is ISO year 2015, Week 1, Day 1.
The Last-Modified header is showing the correct date and time. The Date: header is not.
Last-Modified: Mon, 29 Dec 2014 00:59:30 GMT Date: Mon, 29 Dec 2015 00:59:30 UTC
So, they're using the "G" rather than "Y" designator for displaying the date (if C based) As all the other fields are correct, but they are using the ISO Year, rather than Calendar Year. It's a subtle issue, but a rather silly one.
And clients, can probably see that either a) Mon Dec 29, 2015 doesn't exist (invaild date); or is b) Ignore monday, and 2015-12-09 is too far out of range for a new session token.
These cards have been around for at least the last 4 years. They get mailed out to your home address. Employees only, not contractors.
The cards are for residential support (business is separate) The cards give a gateway-blocked phone number that requests your card ID. And then dumps you into tier 2; Usually into Res-Internet.
Needless to say, I'm not discounting that the lobbying arm of the company has added benefits, and have access to much more influential tools. But the Make It Right cards still relies on existing Customer Support infrastructure. There's no room in that particular system to allow for any real exceptions to give preferencial/better service.
If you go to high in the support chain, your problem won't get solved because of triaging, and work-load, and now issues are being managed by Scrum and Project Managers; and thus you'll wait longer. If you go too low in the support chain, they may not have the experience or know how of how do figure out your issue. Hence the cards get you to tier 2. You bypass the easy-fixes (assuming you tried that once all ready), and thus Tier 2 can help you better and start assessing your issue.
If you use the card, and didn't go through Tier 1 first; you'll basically get the same quality of service as the Tier 2 rep needs to basically do all the tier 1 work with you now. Only difference is that the Tier 2 rep has more experience and may have dealt with your particular issue before (which is also the case with Tier 1), and can bypass some questions due to insight.
Comcast is all about not making exceptions; it complicates business and handling. Top Engineering ands VPs are treated the same way as Call Center reps; at least when it comes to all the details of initial pay, vacation, benefits, cards, tools, etc. Now, the Skilled staff get additional items added on, but these are hacked in. To minimize internal costs, means getting everything onto the same systems, no special cases, and nothing un-audited.
In our division, I have seen, usually near end of year when VPs and Execs send out an email asking "Does anyone have any extra Make It Right cards?"... employees are usually more than happy to forfeit their cards to someone higher-ranked.
Tier one service is adequate for 70-80% of the people calling in. of the above calls, the issue is resolved in one call for 95% of the time. It's the deeper problems that require Engineering Insight, or learning customer state, or escalating to what's effectively Tier 5 support, to escalate to Engineers that cause issues. (tier 2 and above get logged; and increase in weight; usually driving bug-fixes and Engineering time)
The issue is more that no-one has figureud out a way to actually enable good Customer Support. This is an ongoing problem and there is no good solution in the wild yet.
State 1: There are only a few visible symptoms, and end-customers usually have no idea what's going on. State 2: There are literally hundreds of systems internally that affect the customer State 3: For these 10-20 symptoms, there are 100,000s of possible problems.
Problem 1: Hiring hundreds of call centre workers for $10/hr, many of whom have little technical background. Problem 2: Trying to teach these people everything about Engineering, IT, Infrastructure, Systems Architecture, Hardware, Interaction issues, Software Service Issues, Billing Systems, Switching Systems, etc.... and not quit becasue they now know more than most Engineers. PRoblem 3: Because Problem 2 never happens, how does the CS agent search for the solution for your particular problem?
You state symptom 1, 2, and 3.
CS does a search, there are 80,000 possible problems.
CS asks you a question to try to limit.
You perform, and answer.
CS enters that in, there are now 50,000 possible problems.
[repeat until there's a reasonable number]
This leads to Problem 4: Users lie, or misinterpret. If they answer any question wrong, or perform an action incorrectly and give a unknowningly false response, that just filtered out their actual problem, and their problem will never get resolved on that call.
Things like "reboot your modem" are good filters, as that eliminates thousands of possible issues if it causes no change. If you don't actually do this (depending on the problem, they would actually send reset signals, and then require you to reboot; many techincally competent people don't reboot when asked, and thus ) a problem which normally would be fixed with a reboot, isn't, simply because the end-user assumed becasue they rebooted before and it did nothing.
Now, if anyone can design a system that allows unskilled end-users, to communicate their issues, and allow unskilled CS workers to search and find the solution, that would make millions.
For people who ask "train them more" As a fully trained CS degree, Engineering degree, and Engineer at Comcast, I would say that I have no insight as to how hundreds of systems interact or data-relays function. Within my realm, there are thosands of things that can go wrong, thousands that should never happen (yet somehow do, possibly becaues of CS reps changing state on an account without realising the impact). I can fix many issues. But bceause I have this breadth of technical skill, understanding, and knowledge -- would I work CustSupport? No.
If you want better Customer Support, figure out how to make it enticing for highly skilled, trained engineers to work phone jobs; and enough of them to support millions of customers.
The story is BS. Every employee at Comcast gets 3 cards a year. The idea is that if you see or hear someone who's having a problem, you can give them a card and they get a better experience.
The number on the card is a single use number. Thus, once used, it's tied to a specific account/issue, and can never be used again. Second, it's only good for Residential services (Business services have separate support numbers and staff) Third, it only bypasses Tier 1 customer support (newly hired users, who are still trying to figure out all the tools, and the problems,; once you're competent enough on enough systems, you can be promoted to Tier 2.) Thus, if you want the same situation, call in to comcast, and immediate ask to speak to their supervision or a Tier 2 rep; or simply BS that your call was dropped while the issue was being escalated, etc.
Fourth, only a small number of employees actually use the cards. There was a drive to try to convince staff to jus give them out to anybody with a problem; even to friends of friends, or to strangers on the train talking about comcast. Just get them out there.
As the cards are basically tied into the Residential Support system, it doesn't help with Retentions, Service Cancellation, or other non Technical issues with your service. Not sure about billing.
I know when I was at Comcast, I didn't use my cards on friends. Someone complained on twitter about their comcast service, I gave them one of my cards. I gave one to a women I met on a flight; and the last I just lost.
Friends I would direct to call and tell them which keywords to use about their problem so that custrep can find the issue and fix it. (since they're basicaly just using a search engine to try to find out which of the 100,000s of issues your symtoms could match to; which leads to basically hundreds of questions to try to narrow it down, if they haven't experienced your particual problem before)
For every box, there's a minimum of 3 IP addresses. Customer Management IP (Internal IPv6 or v4 depending on location) Customer Modem Service IP (The external internet) VoIP IP Management (Intetrnal address)
And then the public WiFi hot-spot has another.
Most I've seen on the technician's roll out tool was 6 IPs assigned to my residential account.
Caveat to the above: I worked in my Silo; and only on my siloed feature-developement stream; for residential services. Much of the above is based on day-to-day communication and comraderie, but not "hands-on" experience. Thus, the more further removed the service and implmentation (Feature -> Project -> Service Class -> Stack Class in the Residential World) The business world, as far as I know is 99% separated and removed.
When I was still working at Comcast, we went through SIGINIFICANT expenditure to ensure that BUSINESS customers DO NOT have any access to RESIDENTIAL services. Thus, BUSINESS clients, tend to have far more restricted set of services on the account. No access to online voice management. No access to residential technological services No access to 90% of compatible cable modems No access to advanced phone features etc.
This is because we provide additional support guarantees, and additional service guarantees.
The residential services (including the mail relay) go through regular development, upgrades, and service improvements.
But, each additional service a business user has access to, increases support costs exponentially as more and more things can go wrong. This includes misusing a service, relying on a "bug" that gets fixed, the regular downtime residential class services experience due to constant technology and stack upgrades that go on 24/7/365.
And when we discover that a Business customer has access to a new service or feature implementation (this sometimes happens due to Engineers not knowing any better and letting all users access it based on essential requirements); we then have to add in checks, and force-block any business users from using it. (Which can cause complains for the small set of adventurous users), in order to keep the support costs down, and to limit the number of items that can break or be misused.
Personally, I would say that technology-wise, infrastructure, feature-set, and "future-development"-wise. Residential customers get 20-30x the features (IP Telephony; SIP Relay), updates (IPv6), Mail (new Mail server infrastructure and regular upgrades); And residential services are always improved due to end-user complaints. The complaints get bubbled up and filtered, and by the time it gets to Engineering, we have a never-ending stream of technical problems to solve; strange edge-cases, which over time force rearchitectures, new logic, etc. NONE OF THIS HAPPENS to Business Users and Business Systems. They are kept static. No new features are developed constantly - the focus is on hardening and cementing current behaviours and increasing reliability of the current system -- including bugs and broken states that Business Users MAY BE relying on. Fixing them would be a breach of the service contract -- because we do not want to make any change that affects behaviour.
For a Business Service to be added. That's a completely unrelated Full Stack division of support, engineers, management, etc. And they have their own criteria, driven by lawyers about support and features. While a new Residential feature can be conceived, and rolled out within a month -- I have seen the same feature get rolled out to Business after 2 years of constant development on their side to fully describe, monitor reliabilty, full support documentation, all potential bugs and misbehaviours, and hardening. Despite it going into general Residential use (mostly) problem free.
Commerical Users 99% don't want anything to change for any reason, as that costs them money to react to the changes. And Comcast knows that. They will avoid change to busines users like the plague, unless it's provable as required new feature that other business services are providing with the same support guarantees; or the engineers can prove without a doubt the reliability and fully document every possible error, bug, and edge case -- which usually ends up with them requiring to start fixing these remotely possible bugs, error and edge cases untill they become a remote possibility of anything happening. And even then, documentation of what is required to fix it is required, in case it does happen and is reported by a business customer -- which will at that point require it to be fixed outright.
> Then you terminate the contract because it's now useless and the conditions you can use it under have changed - you can NO LONGER SEND EMAIL.
This is not a Comcast issue. The statement "No Longer Send Email" is false. He is still able to sent emails. The problem is that Two SPECIFICALLY NAMED RECIPIENTS are CHOOSING NOT to accept them. Google IS accepting, and receiving the email. And I'm sure other businesses, users, and recipients not on a mass-email-host are receiving them just fine.
Thus, Comcast lawyers can very easily say (with support from network engineers, and email support engineers) that, yes, they are holding up their end of the contract. They are providing a static IP. They are allowing servers to run. They are allowing outgoing Email TCP data streams to fully connect, unhindered.
If you send similar messages all the time, then it'll get auto-flagged as spam as significant repeated content.
I've seen this happen with users having large annoying HTML signatures. All their emails suddenly started going to Spam folder (and I was then not receiving important emails from their other-coworkers with whom I was communicating)
Users are assigned 5 IP addresses. Many block lists are not that granular blocking a/32 address.
Thus, with the user's 23.32.69.15 address: If they block 23.32.69.15/31 (.12 to.15) addresses, that would cover 4 IPs. We do not know if he owns all 4, but it would mean his 5th IP may escape the block. If he doesn't have all 4 in that block, then someone else, assigned an adjacent IP could have triggered the block.
If they block 23.32.69.15/30 (.8 to.15) That would cover 8 IPs, a rule which could be triggered by someone unrelated to him that happens to have an ajacent IP address.
It really depends on how granular the block is. I have pretty much never seen anyone block specific IP addresses before in Emal spam prevention. Normally, I only see/25 (128 IP addresses) blocks and rarely/26 (64 IP address blocks). And provable exceptions within those blocks get white-listed. It's much easier on the spam processing filter to minimize the number of potential rules. So, we over-block. And almost never get any complaints. The major commercial IPs are white-listed at the ACCEPT level (may be further down the line be flagged as SPAM)
The unlisted aspect only comes through the SS7(PTSN) or SIP(VOIP/IMS) protocol headers with a flag indicating whether the account is private, in addition to phone number paying for call, phone number to display, phone number originating, etc... -- AND -- this meta-data can change during a call if it was rerouted mid stream, delayed headers, etc. This gets even more complicated for reverse billed numbers (800) where the originating number is XXX, the billing number is YYY, the display number is ZZZ, and sometimes an interlink number ends up in there. (and as we found out last month with our call logs, some numbers have yet another header that contains virtualized/multi-ring which need to be taken into account; lest the "wrong" number be displayed)
Now, legally, we are required to keep the originating number, time stamp, and length of call;
And for billing and interconnect agreements, the billing number as well.
As we internally always have full access to the raw protocol data on the Enigeering side; the legal siphon (done at the switch level) just skims off all the legally required data and stores it in long-term storage (not DB); to handle the GBs of data a day of the minimally required data.
We then have a separate process which takes each session and generates a [display-phone number, timestamp] DB for 90 days of call logs for users to look up (or legal requirement on bills for chargable calls made depending on juristdiction).
Under no circumstances have we ever kept the "is unlisted" status of the call; as it's never been a datum required for any business logic, ever. And when handling millions of calls daily, and relying on switches to read/dump data for secondary systems to process RT is a space and time sensitive process; and thus, only the absolute minimum required is kept to prevent buffer overruns in the data processing phase;
But, as the process is semi-manual to retrieve data for a given time-range I can understand their request to honor "all my metadata" as well. Limited time-ranges as required by law enforcement is easier to obtain:
- fetch the raw hourly dump files for the time range requested
- run the script that goes through the files and formats a CSV output for any matches of the search phone number
- this process takes hours to run for a weeks worth of data as it churns through TBs of text files if it's outside the 90-day "fresh" window that is stored in a more processed state (but not kept as it's a lot of data to store for no company benefit); most requests from law enforcement only request the last 30days of calls; and this particular process is more streamlined.
- it would be entirely unrealistic to do for the lifetime of a given customer.
One point to take away from this, is that many telecom companies have no interest to keep your data. It's expensive, each item of data adds substantial more costs, overhead, and resource to manage it's storage. It also adds significant more liability as now more people have access to it internally; and safeguards and resources must be used to manage it. Which is why the legal information is done automatically at the switching level, and dumped in a non-processed state; processed and stored, and intentionally kept difficult to access. Because we do not want the liability that comes with storing it, or making it easily available to even a subset of internal employees. Each person that has access adds more risk.
Storing users meta data at least in the telecom world -- is not wanted in the slightest, and we only do the absolute minimum to meet government regulations. Sadly, this also implies that with the current state of laws; that the data is not easily accessible, nor is the data in a state that can be released to a private indiviual without substantial legal risk.
Most everywhere I work, reporting is still the top most requirement. Even more so at publically traded companies.
I've had former colleguges make a good living working in the dedicated report-generation area. (Developing reporting tools, creating reports using existing client tools, etc)
But, the use is primarily that of communication, and more so, consistency, of the data generated so that you can see the trends as they happen; and easily share them in email; slides; presentations; and -- more reports to Regulatory agencies.
Dashboards are nice, but they aren't reports. Reports are normally more complex data manipulation and correlation that are composites and manipulations of the data that dashboards provide. There are also many one-offs that are needed to be drawn up, for specific documents, endeavours, and studies. All of which require good reporting tools. And these reporting tools are lacking in most developers systems. But, thankfully, many developers can expose all the raw data streams, processed into something usable; to which, they take all these numbers, plug them into a proper reporting / modelling toolset, and generate the reports required using the proper tool.
Many places don't have a proper reporting/analysis tool; and expect the software to deliver that. This is a failure of either knownig the tools exist, or unwillingness to accept the costs involved in the additional licenses. (and thus leading to just importing the data into Excel, and massaging it there)
Application 1. generates Metrics 2. exports Data 3. imported into Reporting Application 4. worked by Analysts 5. automatically Generate Reports as new data is imported.
Steps 1 and 2 often exist. Many places want the Application to do steps 2-5, which is fundamentally not the domain. And thus led to the development of dashboards and other simple visualizations, which are not proper reports. Introducing companies to dedicated modelling and reporting tools (Quantrix is one used a lot) tend to get them to realize how much better things could be.... which then usually leads them to complain that their applications don't export data in clean, discrete, normalized data sets to which other tools can ingest.
Most of the large companies I've worked at; and the few government offices I've been at have all had very restrictive email policies. Usually, 100-300mb of email space total. 3 month retention, before automatic deletion.
So, on day 91, emails just go away. It's the best interest of the company normally, especially since they don't want to be liable for maintaining and keeping these archives to legal mandates, especially if one just goes missing. (Exchange does sometimes just, well, lose emails seemingly at random)
It doesn't stop the user to copy the email to a local folder on their desktop, but then, it's now a) against policy; b) not backed up, and c) subject to just vanishing as it's on a non-suported envirnomnet that's prone to being reimaged / upgraded.
Because MRIs and CT Scans are expensive, and aren't done for every crazy walk-in. Most of the time, given an hour or two, everything resolves itself as transient.
As opposed to the US where everyone gets every damned test, and procedure because otherwise the patient may sue, it's how the rest of the world keeps medical costs in check. Don't perform unnecessary tests (as determined by the doctor, NOT an insurance company)
It probably didn't help that she was unable to convey her symptoms to the original doctor; as that probably would have caused them to suspect TIA; or it could be that the doctor she saw wasn't versed in it. Doctor's don't know everything, and tend to come to diagnosis related to their experience. Dietician: diet problem; Urologist: Sex-Hormone issue; Endocrinologist: General Hormone Issue GP: stress/lifestyle issue; Oncologist: possibly cancer.
This is why people need to always get second opinions if they are brushed off. If you get two "it's nothing" opinions, you either have something very rare that will cost lots of money to determine what it is, for minimal benefit -- or, it's actually nothing.
The better situation is -- as a multinational company, they will have to obey all the laws of all the countries they operate within.
They can avoid this by closing the international centres and being located fully within the US.
But then, they would have to pay taxes. As they can't offshore their taxes from country to country; subsidiary to subsidiary to reduce liability to 0$.
So. Is less taxes worth more legal obligations -- for them, the answer is probably yes.
Slashdot Mirror
The negative attitude is what stops ideas from flourishing.
You have no complete files on your computer, just random bits and blocks. At some point, the compression techniques will improve to ensure that data can be reconstructed from these disparate pieces.
Bandwidth is transient, you use your bandwidth, and it's just a passive background task that's always going onwards, and as more units come online into this world, the less bandwidth each unit is required. The files and file systems are designed to handle losses, machines and boxes dying. The blocks are spread across multiple devices and are always kept redundant, so that the chances of data becoming lost approaches zero. If the primary source of your target block is gone, route to another storage loaction of that particular block.
It's is not about large companies supplying all the space (albeit, some will) it's about -every- device offering up space, cpu, hardward, and bandwidth to ensure that the flow of information can never be stopped, becuase it's ubiquitous.
It's for the the full mobility of the data everywhere.
Beyond a single tier provider like drop-box, but spread across every infrastructure storage.
And as advances in storgage increase, those would be deployed and handled the increased load.
The vision is to ensure that day-to-day life goes on, but everyone still needs a hard-drive for local offline storage, boot-up, caching, and the like.
Storage media will not be disappearing so long as ubiquitious internet access world-wide does not exist.
Thus, there will be a market, companies will still have infrastructure.
It's about having every device in the world suppling redundance storage space for multiple copies of data.
Terminoligy needs to be fixed.
All Codepoints are 4 bypes
All characters (defined as a single conceptual, and graphical display unit) range from 1 to 6 code-points. (so, 4-24bytes)
Sinhala:
0dc1 0dca 200d 0dbb 0dd3
ZHA VIRAMA ZWJ RA VOWEL-SIGN-II
Combine to form a single displayable character. (Sri) (kinda a fancy item; but different from without the ZWJ which would display two graphemes. (S', and RII)
And Lituanian:
"However, not all abstract characters are encoded as a single Unicode character, and some abstract characters may be represented in Unicode by a sequence of two or more characters. For example, a Latin small letter "i" with an ogonek, a dot above, and an acute accent, which is required in Lithuanian, is represented by the character sequence U+012F, U+0307, U+0301."
And there are many other cases where there is no single code-point to represent a single grapheme.
So for string truncation and line-splitting, (and anything dealing with arabic or indic scripts), you need to never crop in the middle of a codepoint-sequence that defined a single grapheme; or else the visual display is incorrect, or bakamoji (jibbrish).
Characters in Thai are rendered in display-oredr, and not logical order.
so, for example ( mina would be imna) and requires reordering for sorting.
Characters in many Indic languages are still all syllable based.
So, consonants and vowels are encoded separately, and fully interact as a logical graphical character.
Sinhala:
0dc1 0dca 200d 0dbb 0dd3
ZHA VIRAMA ZWJ RA VOWEL-SIGN-II
Combine to form a single displayable character. (Sri)
If you omit the Zero-Width-Joiner, then it displays as two characters, "Sa'" and "Ri."
So, the rendering and display are dependant on the entire grapheme, which is the normal unit of display and truncation.
Otherwise one will be cropping portions of a character on display; and rendering either jibbrish/bakamoji, or unrelated characters/syllables because.
Malay:
0d15 0d4d 0d38 0d3e
KA VIRAMA SA AA
One displayable character.
If you display code-point by code point, the grapheme displayed would changes 4 times.
KA
K'
KSA
KSAA
"However, not all abstract characters are encoded as a single Unicode character, and some abstract characters may be represented in Unicode by a sequence of two or more characters. For example, a Latin small letter "i" with an ogonek, a dot above, and an acute accent, which is required in Lithuanian, is represented by the character sequence U+012F, U+0307, U+0301."
THe arguments above are missing the point of this development.
It's fear that's the root of all evils, and prevention of advancement.
And the fear, however irrational/illogical on your personal scale is the only obstacle of advancement for all.
This isn't just about putting others' files on your computer, in broken encrypted pieces.
It isn't about the legal ramifications of having random unreadable bits on your hard-drive shard.
What this is, is the future of truly unlimited storage for everyone.
By creating a P2P storage solution, it's creating an "Internet" of storage space that everyone in the world can use.
It can render the local hard-drive solution solely a "cache" of files, but all the files, all the items you access will live on across the network.
If it's done correctly, It will allow one to lose your hard-drive completely, and have all your files instantly available.
Available from any computer interface anywherer in the world at any time.
And, depending on your decryption keys, or more specficially, your custom data-access identifier, you can have multiple file-stores, that are independant, and not related in any way. Or even co-mingling.
This has the prospect of leading the future into a truly data-everywhere situation.
The only item that needs to be resolved, is how to make this information publically available after some time.
History is being lost by the Encryption, and the loss of private journals, of private note-writings, and such.
And over time, it is those items that need to be protected and spread across the world to give insight into who you were, and into your thoughts, dreams, and different views on the events of your lifetime.
But, that can be handled after.
By doing this, we can pretty much guarantee that information can never be lost again, (which is different from ever being exposed.)
Which is a good thing.
I got a ticket doing 65 in the left-most lane; Reason: Car coming up behind me had to break.
It's an enforcable law in most states, that left lane is for active passing only.
More so in Utah, apparently.
They are using ISO Year for the Date header, for some reason. (the last 3 years wouldn't have been affected)
As Mon Dec 29, 2014, is ISO year 2015, Week 1, Day 1.
The Last-Modified header is showing the correct date and time.
The Date: header is not.
Last-Modified: Mon, 29 Dec 2014 00:59:30 GMT
Date: Mon, 29 Dec 2015 00:59:30 UTC
So, they're using the "G" rather than "Y" designator for displaying the date (if C based)
As all the other fields are correct, but they are using the ISO Year, rather than Calendar Year.
It's a subtle issue, but a rather silly one.
And clients, can probably see that either a) Mon Dec 29, 2015 doesn't exist (invaild date); or is b) Ignore monday, and 2015-12-09 is too far out of range for a new session token.
Full Time Employees only, not contractors.
These cards have been around for at least the last 4 years.
They get mailed out to your home address.
Employees only, not contractors.
The cards are for residential support (business is separate)
The cards give a gateway-blocked phone number that requests your card ID.
And then dumps you into tier 2; Usually into Res-Internet.
Needless to say, I'm not discounting that the lobbying arm of the company has added benefits, and have access to much more influential tools.
But the Make It Right cards still relies on existing Customer Support infrastructure.
There's no room in that particular system to allow for any real exceptions to give preferencial/better service.
If you go to high in the support chain, your problem won't get solved because of triaging, and work-load, and now issues are being managed by Scrum and Project Managers; and thus you'll wait longer.
If you go too low in the support chain, they may not have the experience or know how of how do figure out your issue.
Hence the cards get you to tier 2. You bypass the easy-fixes (assuming you tried that once all ready), and thus Tier 2 can help you better and start assessing your issue.
If you use the card, and didn't go through Tier 1 first; you'll basically get the same quality of service as the Tier 2 rep needs to basically do all the tier 1 work with you now. Only difference is that the Tier 2 rep has more experience and may have dealt with your particular issue before (which is also the case with Tier 1), and can bypass some questions due to insight.
Comcast is all about not making exceptions; it complicates business and handling.
Top Engineering ands VPs are treated the same way as Call Center reps; at least when it comes to all the details of initial pay, vacation, benefits, cards, tools, etc.
Now, the Skilled staff get additional items added on, but these are hacked in.
To minimize internal costs, means getting everything onto the same systems, no special cases, and nothing un-audited.
In our division, I have seen, usually near end of year when VPs and Execs send out an email asking "Does anyone have any extra Make It Right cards?" ... employees are usually more than happy to forfeit their cards to someone higher-ranked.
Tier one service is adequate for 70-80% of the people calling in.
of the above calls, the issue is resolved in one call for 95% of the time.
It's the deeper problems that require Engineering Insight, or learning customer state, or escalating to what's effectively Tier 5 support, to escalate to Engineers that cause issues. (tier 2 and above get logged; and increase in weight; usually driving bug-fixes and Engineering time)
The issue is more that no-one has figureud out a way to actually enable good Customer Support.
This is an ongoing problem and there is no good solution in the wild yet.
State 1: There are only a few visible symptoms, and end-customers usually have no idea what's going on.
State 2: There are literally hundreds of systems internally that affect the customer
State 3: For these 10-20 symptoms, there are 100,000s of possible problems.
Problem 1: Hiring hundreds of call centre workers for $10/hr, many of whom have little technical background.
Problem 2: Trying to teach these people everything about Engineering, IT, Infrastructure, Systems Architecture, Hardware, Interaction issues, Software Service Issues, Billing Systems, Switching Systems, etc.... and not quit becasue they now know more than most Engineers.
PRoblem 3: Because Problem 2 never happens, how does the CS agent search for the solution for your particular problem?
You state symptom 1, 2, and 3.
CS does a search, there are 80,000 possible problems.
CS asks you a question to try to limit.
You perform, and answer.
CS enters that in, there are now 50,000 possible problems.
[repeat until there's a reasonable number]
This leads to Problem 4: Users lie, or misinterpret. If they answer any question wrong, or perform an action incorrectly and give a unknowningly false response, that just filtered out their actual problem, and their problem will never get resolved on that call.
Things like "reboot your modem" are good filters, as that eliminates thousands of possible issues if it causes no change. If you don't actually do this (depending on the problem, they would actually send reset signals, and then require you to reboot; many techincally competent people don't reboot when asked, and thus ) a problem which normally would be fixed with a reboot, isn't, simply because the end-user assumed becasue they rebooted before and it did nothing.
Now, if anyone can design a system that allows unskilled end-users, to communicate their issues, and allow unskilled CS workers to search and find the solution, that would make millions.
For people who ask "train them more" As a fully trained CS degree, Engineering degree, and Engineer at Comcast, I would say that I have no insight as to how hundreds of systems interact or data-relays function. Within my realm, there are thosands of things that can go wrong, thousands that should never happen (yet somehow do, possibly becaues of CS reps changing state on an account without realising the impact). I can fix many issues. But bceause I have this breadth of technical skill, understanding, and knowledge -- would I work CustSupport? No.
If you want better Customer Support, figure out how to make it enticing for highly skilled, trained engineers to work phone jobs; and enough of them to support millions of customers.
The story is BS.
Every employee at Comcast gets 3 cards a year.
The idea is that if you see or hear someone who's having a problem, you can give them a card and they get a better experience.
The number on the card is a single use number. Thus, once used, it's tied to a specific account/issue, and can never be used again.
Second, it's only good for Residential services (Business services have separate support numbers and staff)
Third, it only bypasses Tier 1 customer support (newly hired users, who are still trying to figure out all the tools, and the problems,; once you're competent enough on enough systems, you can be promoted to Tier 2.)
Thus, if you want the same situation, call in to comcast, and immediate ask to speak to their supervision or a Tier 2 rep; or simply BS that your call was dropped while the issue was being escalated, etc.
Fourth, only a small number of employees actually use the cards. There was a drive to try to convince staff to jus give them out to anybody with a problem; even to friends of friends, or to strangers on the train talking about comcast. Just get them out there.
As the cards are basically tied into the Residential Support system, it doesn't help with Retentions, Service Cancellation, or other non Technical issues with your service. Not sure about billing.
I know when I was at Comcast, I didn't use my cards on friends. Someone complained on twitter about their comcast service, I gave them one of my cards. I gave one to a women I met on a flight; and the last I just lost.
Friends I would direct to call and tell them which keywords to use about their problem so that custrep can find the issue and fix it. (since they're basicaly just using a search engine to try to find out which of the 100,000s of issues your symtoms could match to; which leads to basically hundreds of questions to try to narrow it down, if they haven't experienced your particual problem before)
For every box, there's a minimum of 3 IP addresses.
Customer Management IP (Internal IPv6 or v4 depending on location)
Customer Modem Service IP (The external internet)
VoIP IP Management (Intetrnal address)
And then the public WiFi hot-spot has another.
Most I've seen on the technician's roll out tool was 6 IPs assigned to my residential account.
Caveat to the above: I worked in my Silo; and only on my siloed feature-developement stream; for residential services. Much of the above is based on day-to-day communication and comraderie, but not "hands-on" experience. Thus, the more further removed the service and implmentation (Feature -> Project -> Service Class -> Stack Class in the Residential World) The business world, as far as I know is 99% separated and removed.
When I was still working at Comcast, we went through SIGINIFICANT expenditure to ensure that BUSINESS customers DO NOT have any access to RESIDENTIAL services.
Thus, BUSINESS clients, tend to have far more restricted set of services on the account.
No access to online voice management.
No access to residential technological services
No access to 90% of compatible cable modems
No access to advanced phone features
etc.
This is because we provide additional support guarantees, and additional service guarantees.
The residential services (including the mail relay) go through regular development, upgrades, and service improvements.
But, each additional service a business user has access to, increases support costs exponentially as more and more things can go wrong. This includes misusing a service, relying on a "bug" that gets fixed, the regular downtime residential class services experience due to constant technology and stack upgrades that go on 24/7/365.
And when we discover that a Business customer has access to a new service or feature implementation (this sometimes happens due to Engineers not knowing any better and letting all users access it based on essential requirements); we then have to add in checks, and force-block any business users from using it. (Which can cause complains for the small set of adventurous users), in order to keep the support costs down, and to limit the number of items that can break or be misused.
Personally, I would say that technology-wise, infrastructure, feature-set, and "future-development"-wise. Residential customers get 20-30x the features (IP Telephony; SIP Relay), updates (IPv6), Mail (new Mail server infrastructure and regular upgrades); And residential services are always improved due to end-user complaints. The complaints get bubbled up and filtered, and by the time it gets to Engineering, we have a never-ending stream of technical problems to solve; strange edge-cases, which over time force rearchitectures, new logic, etc. NONE OF THIS HAPPENS to Business Users and Business Systems. They are kept static. No new features are developed constantly - the focus is on hardening and cementing current behaviours and increasing reliability of the current system -- including bugs and broken states that Business Users MAY BE relying on. Fixing them would be a breach of the service contract -- because we do not want to make any change that affects behaviour.
For a Business Service to be added. That's a completely unrelated Full Stack division of support, engineers, management, etc. And they have their own criteria, driven by lawyers about support and features. While a new Residential feature can be conceived, and rolled out within a month -- I have seen the same feature get rolled out to Business after 2 years of constant development on their side to fully describe, monitor reliabilty, full support documentation, all potential bugs and misbehaviours, and hardening. Despite it going into general Residential use (mostly) problem free.
Commerical Users 99% don't want anything to change for any reason, as that costs them money to react to the changes.
And Comcast knows that. They will avoid change to busines users like the plague, unless it's provable as required new feature that other business services are providing with the same support guarantees; or the engineers can prove without a doubt the reliability and fully document every possible error, bug, and edge case -- which usually ends up with them requiring to start fixing these remotely possible bugs, error and edge cases untill they become a remote possibility of anything happening. And even then, documentation of what is required to fix it is required, in case it does happen and is reported by a business customer -- which will at that point require it to be fixed outright.
> Then you terminate the contract because it's now useless and the conditions you can use it under have changed - you can NO LONGER SEND EMAIL.
This is not a Comcast issue.
The statement "No Longer Send Email" is false. He is still able to sent emails.
The problem is that Two SPECIFICALLY NAMED RECIPIENTS are CHOOSING NOT to accept them.
Google IS accepting, and receiving the email.
And I'm sure other businesses, users, and recipients not on a mass-email-host are receiving them just fine.
Thus, Comcast lawyers can very easily say (with support from network engineers, and email support engineers) that, yes, they are holding up their end of the contract. They are providing a static IP. They are allowing servers to run. They are allowing outgoing Email TCP data streams to fully connect, unhindered.
Gmail filters are also heavily content based.
If you send similar messages all the time, then it'll get auto-flagged as spam as significant repeated content.
I've seen this happen with users having large annoying HTML signatures. All their emails suddenly started going to Spam folder (and I was then not receiving important emails from their other-coworkers with whom I was communicating)
Users are assigned 5 IP addresses. /32 address.
Many block lists are not that granular blocking a
Thus, with the user's 23.32.69.15 address: /31 (.12 to .15) addresses, that would cover 4 IPs. We do not know if he owns all 4, but it would mean his 5th IP may escape the block. If he doesn't have all 4 in that block, then someone else, assigned an adjacent IP could have triggered the block.
If they block 23.32.69.15
If they block 23.32.69.15/30 (.8 to .15) That would cover 8 IPs, a rule which could be triggered by someone unrelated to him that happens to have an ajacent IP address.
It really depends on how granular the block is. /25 (128 IP addresses) blocks and rarely /26 (64 IP address blocks). And provable exceptions within those blocks get white-listed.
I have pretty much never seen anyone block specific IP addresses before in Emal spam prevention.
Normally, I only see
It's much easier on the spam processing filter to minimize the number of potential rules. So, we over-block. And almost never get any complaints. The major commercial IPs are white-listed at the ACCEPT level (may be further down the line be flagged as SPAM)
The unlisted aspect only comes through the SS7(PTSN) or SIP(VOIP/IMS) protocol headers with a flag indicating whether the account is private, in addition to phone number paying for call, phone number to display, phone number originating, etc... -- AND -- this meta-data can change during a call if it was rerouted mid stream, delayed headers, etc. This gets even more complicated for reverse billed numbers (800) where the originating number is XXX, the billing number is YYY, the display number is ZZZ, and sometimes an interlink number ends up in there. (and as we found out last month with our call logs, some numbers have yet another header that contains virtualized/multi-ring which need to be taken into account; lest the "wrong" number be displayed)
Now, legally, we are required to keep the originating number, time stamp, and length of call;
And for billing and interconnect agreements, the billing number as well.
As we internally always have full access to the raw protocol data on the Enigeering side; the legal siphon (done at the switch level) just skims off all the legally required data and stores it in long-term storage (not DB); to handle the GBs of data a day of the minimally required data.
We then have a separate process which takes each session and generates a [display-phone number, timestamp] DB for 90 days of call logs for users to look up (or legal requirement on bills for chargable calls made depending on juristdiction).
Under no circumstances have we ever kept the "is unlisted" status of the call; as it's never been a datum required for any business logic, ever.
And when handling millions of calls daily, and relying on switches to read/dump data for secondary systems to process RT is a space and time sensitive process; and thus, only the absolute minimum required is kept to prevent buffer overruns in the data processing phase;
But, as the process is semi-manual to retrieve data for a given time-range I can understand their request to honor "all my metadata" as well.
Limited time-ranges as required by law enforcement is easier to obtain:
- fetch the raw hourly dump files for the time range requested
- run the script that goes through the files and formats a CSV output for any matches of the search phone number
- this process takes hours to run for a weeks worth of data as it churns through TBs of text files if it's outside the 90-day "fresh" window that is stored in a more processed state (but not kept as it's a lot of data to store for no company benefit); most requests from law enforcement only request the last 30days of calls; and this particular process is more streamlined.
- it would be entirely unrealistic to do for the lifetime of a given customer.
One point to take away from this, is that many telecom companies have no interest to keep your data. It's expensive, each item of data adds substantial more costs, overhead, and resource to manage it's storage. It also adds significant more liability as now more people have access to it internally; and safeguards and resources must be used to manage it. Which is why the legal information is done automatically at the switching level, and dumped in a non-processed state; processed and stored, and intentionally kept difficult to access. Because we do not want the liability that comes with storing it, or making it easily available to even a subset of internal employees. Each person that has access adds more risk.
Storing users meta data at least in the telecom world -- is not wanted in the slightest, and we only do the absolute minimum to meet government regulations. Sadly, this also implies that with the current state of laws; that the data is not easily accessible, nor is the data in a state that can be released to a private indiviual without substantial legal risk.
Most everywhere I work, reporting is still the top most requirement. Even more so at publically traded companies.
I've had former colleguges make a good living working in the dedicated report-generation area. (Developing reporting tools, creating reports using existing client tools, etc)
But, the use is primarily that of communication, and more so, consistency, of the data generated so that you can see the trends as they happen; and easily share them in email; slides; presentations; and -- more reports to Regulatory agencies.
Dashboards are nice, but they aren't reports.
Reports are normally more complex data manipulation and correlation that are composites and manipulations of the data that dashboards provide.
There are also many one-offs that are needed to be drawn up, for specific documents, endeavours, and studies.
All of which require good reporting tools.
And these reporting tools are lacking in most developers systems.
But, thankfully, many developers can expose all the raw data streams, processed into something usable; to which, they take all these numbers, plug them into a proper reporting / modelling toolset, and generate the reports required using the proper tool.
Many places don't have a proper reporting/analysis tool; and expect the software to deliver that. This is a failure of either knownig the tools exist, or unwillingness to accept the costs involved in the additional licenses. (and thus leading to just importing the data into Excel, and massaging it there)
Application
1. generates Metrics
2. exports Data
3. imported into Reporting Application
4. worked by Analysts
5. automatically Generate Reports as new data is imported.
Steps 1 and 2 often exist. ... which then usually leads them to complain that their applications don't export data in clean, discrete, normalized data sets to which other tools can ingest.
Many places want the Application to do steps 2-5, which is fundamentally not the domain.
And thus led to the development of dashboards and other simple visualizations, which are not proper reports.
Introducing companies to dedicated modelling and reporting tools (Quantrix is one used a lot) tend to get them to realize how much better things could be.
Most of the large companies I've worked at; and the few government offices I've been at have all had very restrictive email policies.
Usually, 100-300mb of email space total.
3 month retention, before automatic deletion.
So, on day 91, emails just go away.
It's the best interest of the company normally, especially since they don't want to be liable for maintaining and keeping these archives to legal mandates, especially if one just goes missing. (Exchange does sometimes just, well, lose emails seemingly at random)
It doesn't stop the user to copy the email to a local folder on their desktop, but then, it's now a) against policy; b) not backed up, and c) subject to just vanishing as it's on a non-suported envirnomnet that's prone to being reimaged / upgraded.
Because MRIs and CT Scans are expensive, and aren't done for every crazy walk-in.
Most of the time, given an hour or two, everything resolves itself as transient.
As opposed to the US where everyone gets every damned test, and procedure because otherwise the patient may sue, it's how the rest of the world keeps medical costs in check. Don't perform unnecessary tests (as determined by the doctor, NOT an insurance company)
It probably didn't help that she was unable to convey her symptoms to the original doctor; as that probably would have caused them to suspect TIA; or it could be that the doctor she saw wasn't versed in it. Doctor's don't know everything, and tend to come to diagnosis related to their experience. Dietician: diet problem; Urologist: Sex-Hormone issue; Endocrinologist: General Hormone Issue GP: stress/lifestyle issue; Oncologist: possibly cancer.
This is why people need to always get second opinions if they are brushed off. If you get two "it's nothing" opinions, you either have something very rare that will cost lots of money to determine what it is, for minimal benefit -- or, it's actually nothing.
The better situation is -- as a multinational company, they will have to obey all the laws of all the countries they operate within.
They can avoid this by closing the international centres and being located fully within the US.
But then, they would have to pay taxes. As they can't offshore their taxes from country to country; subsidiary to subsidiary to reduce liability to 0$.
So. Is less taxes worth more legal obligations -- for them, the answer is probably yes.