Yep but to arbitrarily set a time would end up causing back and forth between ACCC and the Manufacturers about how long is the right amount. I was just suggesting a model that is already commonly used to define that 3 year time frame. I really would like to see even longer than that ie 5 years but there currently is no common model that is in evidence and that supports that period.
My main concern with setting hard time rules is that manufacturers will adhere to that with some tenacity and the consumers who suffer something wrong after that period no matter how much they expected it to last longer will be left almost without recourse. In this regard I like the ACL for it’s flexibility to address issues that become apparent long after any warranty has expired eg Takata airbags in cars. Manufacturers can and do set a time limit but the ACL overrides that. The ACCC also embrace the extra benefits when express warranties or a statement of quality are offered to no disadvantage of the consumer as they just become legally binding additional protections that do not override the existing ACL rights.
Microsoft has just issued a patch for Windows XP, five years after it ceased support, because of a major vulnerability in its Remote Desktop Protocol (RDP - never turn it on). That is an exception.
Should Intel recall and replace every single CPU it has sold that uses speculative execution? That’s eight years (at least) of products that are all affected by the same problem. Most users will never have to worry about it, though, because we don’t have strangers running things on our machines. Then there’s Cisco, which has something that has been named Thrangrycat (you’ll have to visit the site to understand how truly awful that name is) which affects millions of Internet-facing devices!
IT security is a moving target. Intel introduced speculative execution for performance benefits, and several years later researchers discovered that it could leak sensitive information. Have a look at the monthly Windows patch updates - I think there were 70+ fixes in the latest one, while Adobe synchronises its patches with Windows and releases similar numbers. Monthly.
There are millions of people world-wide developing billions of lines of software every month, and there are going to be many many flaws in that code. Similarly, hardware developers create firmware (and even hardware itself) that is hopefully bug-free but reality proves that nothing is bug-free.
Please, no. That legislation is so abominable it should never be mention in impolite company.
Not if the feature’s in hardware and not if it’s prime functionality (such as with the above-mentioned Cisco routers - although patches are available for them).
All of this ignores the elephant in the room, though. A manufacturer can design a product they think is secure, release it, learn about bugs and release patches. They can set the product to auto-update (as Windows has done in its last few iterations). A fairly large proportion of the product’s owners will either disable the auto-update because their company needs to test every piece of new hardware/software before it’s deployed, or not even know they have to reboot the device because it just sits in the corner and runs - year after year - without complaint. It’s downloaded an update, but the user doesn’t know because the device has no way to tell them. These are the devices that are making up today’s botnets, and making the Internet a battle-ground. It’s not the educated consumer that’s the problem, it’s the overly paranoid enterprise and the consumer who plugs it in and expects it to just work forever.
(And I suspect that we have strayed somewhat from the original topic of this thread.)
They would first need to have CPUs that don’t have any of that class of vulnerability (or the vulnerabilities in that class that have been made known to them so far). LOL.
Some of us are throwing around a mandatory 3 year warranty (not 8 years) so that would limit the size of the recall.
The real problem with a recall of a CPU is that you can’t necessarily plug a later working CPU in at all. It is not necessarily a drop in replacement. So where that is the case that means either replacing the whole motherboard or backporting the fix in silicon to the earlier generation of CPUs. Yuck!
But why not make manufacturers who make zillions of dollars of profit accountable for the defects in their products?
Interesting vulnerability. They could at least fix the “remote command injection vulnerability”. That should make an exploit that requires local access in which case the would-be intruder has many other exploit options.
Sure. Noone is saying otherwise. We are saying that when vulnerabilities are discovered the manufacturer must take steps to fix (for more than 1 year).
I share your disgust. However it exists so it might as well be used for something positive.
There will be times when this is 100% right but there will be other times when allowing something that is broken to be disabled by the customer is both possible and a benefit to the customer.
Unfortunately quite a few naive pedestrian users do have strangers’ code. They are under the general headers of viruses and bots. OK, not most, but quite a few.
Windows 10 1903 has added an icon to tell the user a reboot is needed to install an update. Want to guess whether it can be turned off? No need to guess
and that assuming updates are enabled, and Microsoft is increasingly proactive about pushing them, at least onto the consumer installations without corporate deployment/control.
(This is in the Updates->Advanced Options page - if it was previously there I never noticed it)
sandboxed code also falls into this category. Javascript is the most widely used, although by now most browsers have been patched to make this exploit harder to carry out in Javascript. Java has the same issue, although many browsers outright disable Java these days.
A washing machine has raised its hand! The advertisement leaves functionality to the imagination, but the obvious is an alarm when a cycle finishes or gets interrupted, or maybe the more conscientious can keep track of a cycle in real time
Hi all, I have a synology rt1900ac router, courtesy of a suggestion from this site many years ago. So now I’m back and I would love some advice regarding mesh or extenders? We have a few devices and I would love to keep my current synology router, as it’s functionality fits well. But, I want something that will grow with home automation and also extend the current wifi signal beyond 2 rooms.
I’ve used both range WiFi range extenders and ethernet cabling to provide coverage in larger homes.
Experience with WiFi is that capacity and range very soon become a constraint. Not so much an issue if you are on an under performing NBN connection. Experience with a basic NAS and later a higher speed raid based storage led me to install copper to connect between the distributed wireless access points and router through a gigabit switch. Also to connect other devices directly to the home network.
If your needs are modest a WiFi mesh network may be all you need.
If you are well settled and need lots of bandwidth now in the future it may suit to have some cabling installed to reach 3 or 4 locations around the home.
If you can share a few more thoughts on what you think you need your network to deliver it could assist others to offer their thoughts/experience.
Footnote.
On your existing 1900AC device, it’s important to check that it is still receiving firmware and software updates. If not it may be vulnerable to various exploits, in particular when used as your primary internet connection device. Most devices seem to loose support within 2-3 years of being superseded.
I have a mix of mesh, access point, and extender, and it works well.
Our former modem Netgear modem/router was deemed obsolete when we were forced onto HFC NBN. NBN provided a modem, and I attached a Google Nest Wi-Fi mesh router to the modem, with a gigabit switch hanging off that.
There was some legacy Cat4 UTP going to another floor from the gigabit switch, so I turned another Netgear Wi-Fi router into an access point using the same SSID. Further away, I located two Google mesh Wi-Fi points which only link up with the Google router. Further away again, I installed a Wi-Fi extender. It sounds crazy, but this hotch-pots rig gives us seamless high speed wi-fi coverage over a large area, and through many layers of obstruction using only a single SSID.
The point is that you can meld your legacy equipment in with modern automation provided by a mesh.
The best type of devices to use to extend you home network in my view, is a Gigabit Ethernet & Wireless (Wi-Fi) 802.11Ac Powerline Adapter which consists of 2 separate units. They are referred to Ethernet Over Power (EOP). The base unit plugs into a 3 pin wall socket & is connected to your router, (any router) via a short Ethernet cable. The remote unit plugs into any 3 pin wall socket situated, say, about in the middle of the house. It not only has an Ethernet socket to connect to devices via an Ethernet cable, but it also has high speed Wi-Fi signals, for a number of devices to be connected wirelessly at the same time.
Range extenders are next to useless in my view after research, for various technical reasons, and the above described units send the Ethernet signals at near Ethernet cable speeds, over the mains power-lines in your house. The receiver unit can be as far away as 300 metres from the main unit. (yes 300 metres) as long as it is connected to the same wiring system.
The units are not cheap & from about $180 to $200. They are well worth it though, & much cheaper option than Mesh networking modules, or paying for Ethernet cabling to be installed.
If you are interested in the above, you MUST buy one that has Wi-Fi capability on the remote unit, as some of the suppliers use different and confusing terminology to describe their units, and they may not have Wi-Fi capability, otherwise you can’t connect multiple devices at one time using the Ethernet socket at the remote unit.
Here is a link chosen randomly to a supplier that I have no relationship to or pecuniary interest in. Its just to show you what a suitable unit looks like. https://www.mwave.com.au/product/netgear-plw1000-gigabit-ethernet-wireless-80211ac-powerline-adapter-kit-ab94886?gclid=EAIaIQobChMIxIqJvdzZ7wIVWn8rCh3v1ARMEAQYFCABEgISM_D_BwE
Hope this helps you
The gotcha with those is both units need to be on the same electric circuit in the house. If one is on another breaker there is a Rafferty’s chance it will work properly or at all. Success varies with the specific Powerline adapter product and one’s home wiring.
I have a D-Link DHP309AV for a problem TV (OT for this topic) and both adapters are on the same circuit (breaker) and work a treat.
As a still registered, but retired, A-Grade Electrician, I find it hard to fathom why the remote unit wouldn’t work as well off the different circuit, than the one the base unit is connected to, but since you have experienced this issue it demonstrates that it is still possible. It maybe possible, if one or the other of the circuits is fitted with a safety switch (RCD), as opposed to a regular circuit breaker, that this could affect the unit’s operating efficiency & effectiveness.
However, an electrician would be easily able to re-arrange the circuitry at the switchboard, to make both units on the same circuit breaker.
Some will degrade and others won’t even work if you plug one into a power strip or extension cord. That and the single electrical circuit are documented in the owners manuals for Powerline products. Some make it clearer than others. From my DHP-309AV manual
My house has a front and a rear circuit. It would be fairly complex to move powerpoints from one to the other. $$$$.
As I wrote it is a Rafferty’s issue and the only way to know is to try it, but a consumer has to make sure they ask ‘the question’ when purchasing so if it does not work, it is not a change of mind, it was being sold an unsuitable product. To be clear some are reportedly far more capable products than others.
We thought the same thing…but I understand it depends on how the switchboard is wired and the distances from the powerpoint to the meterboard and back to the other powerpoint. We had a house in Brisbane where we would run some TP-Link powerlink adapters over two different circuits (one in the old house and a second circuit in an extension). They can work, but as @Airsie has indicated, it might be dependent on how the switchboard is wired.
Phil that is a very good, and important point that you made about not connecting Powerline Adapters via a power-board, extension lead, or a surge protected power board, that I overlooked mentioning.
The Powerline adapters must ONLY be plugged in to a permanently wired wall socket outlet.
The use of an extension cord with a powerline adapter is possible. It will usually have some degraded performance over plugging into a powerpoint but it may just be minimal.
Reasons why it may affect performance are largely due to:
Interference from the increased length of the cord (and the quality of the wire used eg gauge).
Some cords have built in anti-surge and noise filtering (not many here)
If the cord has any looping this will increase noise substantially and cause a lot of loss of signal
So if you absolutely need to use a cord make sure it is as short as possible, avoid loops, avoid passing the cord over other cables eg ethernet ones or other power cords, and ensure it has no built in filtering or anti-surge (unlikely here but some do have built in lighting which can add to noise). I am using one as the powerpoint is at the side of an Entertainment Unit and the powerline brick is too thick to fit into the powerpoint because of the Unit. I am using a flat headed cord where the cable comes out of the side rather than out of the center of the top. The powerline has a passthrough which a powerboard is plugged into. This is for the TV as I prefer wired ethernet over WiFi. Rock solid Net connection.
Powerboards can also be used. They often have anti-surge, and filtering built in. Don’t use the surge protected and or filtered ones. Ensure it has a proper earth connection as sometimes they only use two pins out of the three and for some reason this seems to affect transmission, also keep EMR noisy items away eg don’t plug in a hair dryer into the same powerboard. BT (British Telecom) have a page about using powerboards with powerline devices
These all work from my personal knowledge/experience, taking some care about choices of the products such as cord means I have little loss on the line. You can search online and find many who use them, some even piggybacking powerboards together and still getting reasonable rates of transmission. I haven’t ever piggybacked so I can’t say how much effect it has.
I’ve used a pair of Dlink branded Power Line ethernet adaptors for several years to link the TV directly to the internet/router. It’s cheap and easy to use an Ethernet switch to increase the number of ports and physical devices able to connect to an adaptor. I prefer to connect storage, printers etc directly to the network via Ethernet cables. It’s been more consistent than relying on WiFi.
Note:
On the maximum connection distance of 300m @Airsie mentioned. It’s the theoretical maximum length of cable between two or any devices. In practice due to cable losses (attenuation), the practical distance is much less. Don’t expect it to be an automatic solution to connect the farm shed 250m away. The cable run is likely longer than line of sight.
Our home has 4 distinct power circuits as it is built in 3 portions, with each circuit protected by seperate RCBO (integrated circuit breaker RCD switches). These can each cause up to 50% signal loss when the adaptors connect between circuits. Twice that across any two circuits. The Ethernet adaptors use signal frequencies up to 86MHz in one specification, hence they are subject to RF losses and interference.
Last night I discovered another anomaly, noting I have only had the powerlines for a short time and continue to learn about their foibles.
When the DLinks go into power saving mode the ‘offending TV’, a Sony 49X8000H (2020 model) will neither wake them up, nor does it then recognise there is anything in its Ethernet port anymore. While awake the DLinks work perfectly. As soon as they go to sleep it goes pear shape. A quick on/off of the power wakes the Dlinks and ‘fixes’ it, but that was not an expected ‘feature’.
The wonderful world of IoT and how they interact with powerline adapters is, like the characteristics of the powerline adapters, very product dependent.
edit: Continuing with trial and error and with too many insights about how engineers and software people [don’t always] think I tried a static IP in lieu of DHCP. The Sony then knows there is something plugged into the ethernet port whether or not ‘connected’ to a network; the first network access fails but a retry wakes the DLinks and does the job.
Would you find a mains power run that long on your property? My understanding is that you would get noticeable voltage drop over such a distance. Out where each property has its own transformer one is advised to keep 230 v lines under 100 m and preferably under 50.
