Posts by iced.lemonade 38 publicly visible posts • joined 16 Aug 2016
i feel the research resource may be better spent on some crucial areas, such as (1) improving the capacity, (2) improving the reliability and (3) improving the safety of the batteries, rather than some nice-to-have areas like (in us) bidirectional charging or (in europe) in-car entertainment...
otherwise the ev car industry may be dominated by the asian auto makers in a not-too-far-future - just see how the battery technology is evolving in mainland china.
AI after all can only process what is being fed in, it only excels at combining previous data in unexpected ways - we can take that as inspiration, but not as a source of knowledge. For example, when we are out of ideas or want a completely new view of issues at hand when we have reached a bottleneck in the thinking process, AI can help, but its output - every time - needs to be proofed and verified with care. Maybe there will be an increase in some kind of DSLLM (domain-specific LLM) which focus on one domain of knowledge, it will certainly be more 'intelligent' in some sense - just like human where we become experts when we starts to focus on some specific type of knowledge and research, instead of the huge LLMs which tries to be a be-all type of knowledge machine.
It seems that bigger model introduces bigger chance of ambiguity and that cannot be avoided, and ambiguity itself is a limiting factor to the usefulness of the models, other than for inspiration.
if apple can do a remote desktop thing to other macs (for example, an m1 air remoting a mac studio in office) that is as performant as windows remote desktop (rdp) then many people will switch to macs, i think. the speed, the latency and the power consumption of m-series is great, but lack of a performant remote desktop solution is... a deal breaker (for me at least), and vnc or even anydesk is no match for rdp.
perhaps apple will speed up thing by putting some remoting-related code into kernel to truly speed things up?
i think the pc's still have the edge of supporting large and upgradable memory, flexible graphics options and upgradable storage, but for users that don't need these 3 m-series may already satisfy all their needs (like home users who aren't gamers, office workers, etc.).
and i hate it when windows updates breaks my printing functions almost every time... for most of the time removing/re-installing drivers (and sometimes several steps of additional configuration) will correct the problem, but why does me need to wipe the floor after ms throws some sh*t on it? i paid for the product and support their telemetries, after all...
I work for an SME and with limited headcount here it seems not quite feasible to replace all stuff written in Java here... and after several attempts at replacing Java with alternative technologies (like golang, rust, etc.) we found that we just keep coming back to Java, as we meet at least one major roadblock for each alternative we evaluated. It is probably due to the fact that our skills are limited, but our man-years in Java-based software development just cannot be replaced with our limited budget and time.
Java is, to us, a language that just works. We just hate the licensing model of Oracle. We were hesitant to use alternative JDKs at first but it has been several years now (for serverside, we are all on Temurin now) and the deployment history has been quite smooth. And we are also evaluating supports from alternative vendors like Azul (their Apple Silicon JDK is awesome) - we fully understand that no software are created freely (as in beer) and all developers need to be paid... we just want to support Java with licensing in a more sensible way. It seems Oracle are in full throttle targeting for the biggest paycheck and we, maybe like other Java-dependent SMEs, will just seek support from other vendors that are more SME friendly.
comments of anti-vaxxers, or those who only reject _mandatory_ jabs, usually get downvoted to hell... not sure why people cannot try to believe that vax is more of protecting people against serious symptoms and less of against transmission (at least this is said by some of the medical doctors here in hong kong).
so, yes, anyone can jab how many times they want to be, well, protected and since it's more of protecting yourself, it's seems a bit far-fetched to force other people to get the jab to... protect themself?
it's seems like in a cold day you wear warm clothes to protecting you against catching flu or whatever, and then you make illegal for people to wear their short sleeve clothes to get outside.
this sort of logic applies equally to the vaccine pass... if you have got enough jabs and you believe the jab works as advertised, you should be less worrisome of getting covid and how come the government need the vax pass to protect... the vaxxed?
anyway, hopefully covid is almost over and both the vaxxed and un-vaxxed can enjoy life together, under the same sky.
libreoffice?
at least it is on-premise and don't send your anything anywhere automatically.
i know it may be a bit poor, feature-wise, in comparison to options by the big corps, but in this case it is academic use so maybe it is enough for the purpose.
or if it cannot fit completely in the academic ecosystem, maybe encourage usage in the school <-> student <-> parent loop or even school <-> school loop... and for the school <-> education authority / government loop just continue to use the big name thing - and replace it with LO if, ever, it becomes feasible.
I am no marine scientist, but i suspect many ocean dwellers (fishes, corals etc.) are sensitive to the temperature of the sea water (maybe big change will happen even if it is +/- 1 degC, like massive death), and if this below-sea data centers proves to work then there will be a rush to put these heaters under the sea(much lower cooling cost! what's not to love!) and a catastrophe will happen to those ocean dwellers and the surrounding ecosystem.
Hope I am wrong.
it's just a more rational approach to giving money to startups... in good economical climate you can be a bit more optimistic but now, when everyone is tightening their belt the valuation of the startups naturally falls as the potential revenue falls across the whole landscape and each dollar is given out with considerably more justification.
a distro designed for desktop usage will never be good enough for server usage, and vice versa. it seems suse is going to be server-friendly so it's a signal to the desktop users to migrate to a distro better for desktop use - there are plenty, like zorin, kubuntu etc.
what i don't understand is the "adaptable" thing... at the first glance, i thought they are making the distro more modular... for example, instead of low-level libraries they group them into modules and allow users to (1) setup, (2) configure, (3) patch and (4) upgrade them at ease. there may also be different "recipes" for system configuration, for example, json files (just 1 .json file for each recipe, please) which configures the modules which can be shared between users and easily applied to their systems. a plus for this is, there will be a lot less distro which differs only in a very minor way - a new distro is just a json file away.
if they want to go the containerized way, please do it more deeply - how about, in addition to each application running in a container, there are "data containers" which, like docker volumes, that share data between containers? for example, application a and application b want to share a .png, and the system creates a new data container just for sharing between a and b (which is invisible to application c)... i have always long for this functionality, but maybe someone have already done it - i just think if someone could make this work in a straightforward way, it would be a big plus to security of a desktop-oriented distro. my mac always ask me if i allow some app to access the "Downloads", "Documents" etc. folders but it is useless as all apps accessing my "Documents" can access everything in it.
I am not really sure why many people hate Snap to that magnitude other than it takes way more resource than a traditional apt-packed application... I feel the desktop is increasing powerful and the web is increasingly hostile and keep desktop app containerized keep your system clean... you know, one bad-behaving app ruins your setup... usually, other than a handy number of application which requires host access (for example, FileZilla) I feel it is safer to isolate each of the app I use.
But, the utopia for me is each application is Jailed and packaged with extension .jailed (like firefox-1.2.3.jailed) so they at least cannot see or touch _data_ which they should not see or touch.
Paid for Zorin 16 and the experience is nice. The UI is snappy and it feels polished and without dumbing everything down. One point for them is their support of Nvidia graphics - they have a boot option to use a better Nvidia driver and that's is the only way that I can use linux with decent graphics on my laptop (it is a MSI with a 2070) - without that the graphics (especially with external monitor) is super slow. I am sure it is doable with other distributions like Ubuntu, but I don't have the persistence to drill into the forums.
It feels good to pay for them (in no way related to their donation to the Ukraine - i will donate directly to the charity there), in a sense that it needs money to develop and support any piece of software (and their's a good one). Hope my minor contribution will be useful for them to continue their effort.
It seems that Mac lineup is now targeting the traditional target audience once again.
Mac Mini, Macbook Air - mainly educational clients;
Macbook Pro, Mac Studio - mainly content creator, like graphic designers and video makers.
The M1 is engineered exclusively with those users in mind - instead of being a general purpose chip that is all-rounded for business people and the like. I am not saying that users wishing for a general purpose machine won't buy them; they are just not M1's target.
Once Apple machines was sold in small shops with a colorful Apple logo, and there were standard pamphlets of the current attractions that is placed at a corner (like, Macintosh Performa, Power Macintosh, etc)... Apple was a much less profitable outfit in those times, but I remember that kind of cult feelings.
stop wasting effort like neutralizing technical terms for inclusiveness (for example, the endless debate of male/female/master/slave...) and start competing. don't assume that you are leading in tech - always have the mindset of an underling. especially after the covid where many people are becoming poorer in general and if you cannot compete dollar per dollar (i am not saying cheaper is better; i mean the worthiness of each dollar spent) the game is lost. don't under-estimate the power of your opponent - workers in china, in many industries, are like robots; they can churn out stuff with much less overhead than others do, especially in terms of lower labour cost, less consideration to the environment, more hazardrous working environment etc. and they copy (ideas, etc.) and feel that's a natural thing to do.
if you want to sell your stuff at higher price, prove that it worth it. sometimes i even wonder if open-source is a good thing - it seems the idea of open-source was bred in a previous era where the world was simpler, and now it feels like giving free weapons to your opponent to shoot at you.
maybe i am biased, but as a hongkonger who are experiencing the increasing aura from mainland i feel it never hurts to be a bit cynical.
after reading much of the comments in this post i was finally inclined to study the ipv6 specifications and features and, yes, it is designed to end the limitation presented in the ipv4 designed for the yesterworld, just like storing dates in 32bit vs 64bit number.
maybe it is that the ipv6, at the first glance, is so foreign, and that unfamiliarity turns away people at the first glance. perhaps more information on the subject to layman (or those work in IT with technical knowledge not much more than a layman) will go a long way for public adoption of it.
i am no network engineer, but if the problem of IPv4 is the scarcity of available address, is it possible to prepend the existing 1.2.3.4 to 0.0.0.0.1.2.3.4, update the IPv4 devices that recognize this extension, and call it IPv4 R2? seems that existing v4 devices, ideally, will recognize a subset of the R2 address pool but can still function until it makes the transition to R2...
my work depends on remote desktop into the workstation at office, and generally rdp is really performant - with my broadband link at home it's responsive. i primarily use various intellij ide and i almost forget that i am actually remoting, instead of working at local machine.
nomachine comes next but when i use ftp on the remote machine it slows down to a point where it is unusable - it seems windows prioritize rdp over other network operations.
then come rest of the remoting solutions, like team viewer, vnc etc., where i find those generally unusable for daily driver. maybe it is my niche use case, but running ide remotely with solutions other than windows rdp is painful in general, as responsiveness is crucial (for example, in code suggestions, where small difference in responsiveness makes a big difference).
then a question: X was designed with non-local access in mind, so it is supposed (or not?) to be at least on par with windows rdp performance?
the strength of any AI systems, ultimately, depends on the amount of data that is used to train it. the more data it churns, the better result it yields (i am no data scientist, but i feel if the amount of data increases beyond a certain amount, even if the data is biased, its effect will become insignificant - please correct me if i am wrong). in china every message sent, including voice and text messages, through any messaging software allowed in china is analyzed by the central government, for the statewide credit system, and given that the credit system is central to the political system in mainland china, naturally it is given unlimited amount of resources. i don't think any individual company can match the amount of data used to train china's AI - well, maybe if facebook, google and microsoft would pool their resource for a common AI initiative that may have a chance but i don't think that will happen anytime soon. i am not saying that the sharing of data between those 3 companies is a good thing, but technically, it takes a huge effort if the western side want to match, let alone outpace, AI development in china especially in the area of language processing.
as someone mentioned above that the menu bar actually bleeds into the previously unoccupied area of the monitor in full screen mode, i think it is kind of odd... every app that supports full screen will have the root menu items cramped into the left half of the menu bar or the menu items will be split into two - most of the menu items will be on the left side and one or two remaining will be on the right side, along with the statuses and what's left.
menu bar is the center of activities of macintosh since the first System, and i feel apple is unknowingly, partially destroying it with that "notch".
with Big Sur apple makes the menu drop-down rounded corner in all 4 corners and i already feel it's unnatural - it feels detached and it is better if the rounded corner only apply to the bottom 2 corners and leave the top 2 corners square (so it is a menu drop-down instead of feeling like a pop-up), but it is just my perfectionist 2 cents.
i hope apple will not introduce Hamburger menu to macOS but... who knows.
the country now protecting citizens' users rights while their human rights are gone.
next they will be protecting citizens' privacy while everything they do are logged and analyzed by government's credit systems.
i feel this gonna be downvoted to hell but i wonder... what does 'protect' mean to them?
i wish someone make a browser like firefox, without google influence, for a subscription - say, $5 a month - just like anyone using vpn, instead of tracking me to the end of the world and sell my path away. i am a software developer myself and i prefer to pay for the effort of the developers rather than let my web access profile sold to third parties that i would rather not to. but maybe our profiles are too valuable that they cannot set a reasonable price tag on their browser to justify the whole thing.
just my wish.
as a primarily backend developer with occasional frontend jobs for years, what dart + flutter appeals to me is the ease of coding the ui / flow of the whole app without ever using GUI. personally i like programmatic approach to ui stuff - i hate the storyboard stuff on iOS development, for one - because of the flexibility and speed in implementing changes between iterations. if your company is large enough to employ dedicated teams for both ui/ux design and front-end developers that may be another story, but in that case you may not be using this lowcode flutter stuff at all because there are much better options out there.
to sum it up, dart + flutter, in my opinion, is designed from the ground up for programmatic approach to ui development.
It takes time to make manufacturing decentralized into other countries - some of those manufacturing functions requiring less technically-capable labour has already spread into SE Asia, and those technically mid-tier are going/gone to India. The 'Global Factory' wasn't built in a day and, perhaps, decentralizing it requires much more time than it was built, especially infrastructure-wise - for example, some countries like Bangladesh have cheap labour but the electricity supply is unstable.
https://sciencenorway.no/covid19/norwegian-experts-say-deadly-blood-clots-were-caused-by-the-astrazeneca-covid-vaccine/1830510
Transparency and openness important to build trust:
https://sciencenorway.no/covid19-vaccines/the-astrazeneca-debate-has-not-created-anti-vaxxers-in-norway/1853379
i think the critical issue about vaccine is that for different people with different health status / age / medical background, there are just so many unknowns on the side-effects of each kind of vaccine which is unconvincing to many people who are not facing severe risk of infection vs much higher risk of undocumented outcomes after taking the jab. if the supplier of the vaccine / health organization providing the jab are more transparent to the result of the ones taken the jab many more people will take the jab because they have the data, can decide to jab or not or which kind of vaccine they will take with justification based on that data. it's ultimately a matter of trust more than anything else.
i'm from hk, and i am taking a variety of blood-sugar-regulating medications and others so i won't risk myself to the vaccine yet. everyone in the gov't here and other countries seems to tone down the consequence if you have any health problems that you may have, and unknown to you and even your family doctor, before taking the jab. for example, in my limited knowledge, people with long-term health problems, like SLE, or other people with potential issues (especially blood-vessel-related) with their heart / brain, and those aged 60 and above, are likely to have fatal or unrecoverable consequences after taking the vaccine, and the 'experts', when facing the report of such cases, routinely say something that 'they already have condition a, b, c and some of x, y, z so they are dying normally and not linked to the vaccine' and it is hardly assuring.
When I first programmed in Java when it was 1.0.2 on my Mac, it was very simple and straightforward - seems most syntax are there for a reason. Then there was varargs, generics, etc. which is still sane additions to the language. But under oracle stewardship, seems they are changing for the sake of changing. I think, syntax-wise, Java 8 has been the apex where even if they don't change the syntax for another 20 years it would still be fine. They should really focus on the standard library and work from there.
I started my career as a coder spending most time on the UI side - and I think the swing and javafx libraries - they need much more love than oracle gives them and it would be more purposeful if oracle work on them instead of adding syntactic sugars which makes everyone diabetic.
It is like working as a data-entry clerk with the keyboard layout changing every year with the benefits that can only be felt by the keyboard maker.
Btw, I think the changes that is really essential at Java 9+ is the module system and the native packaging of Java 16. Other coders may be attached to other parts of the Java language/library but just, slow down the pace of changing the god damn syntax - (1) the more freedom in the ways the user codes, (2) the more styles the application would be coded in and (3) the less readability of the code in the long run. The readability factor alone is justified of using Java instead of C++, in my experience, if oracle would just keep Java simple.
The mass media all over the western world are all covering the news on Biden's side and they don't even bother to report the possible evidence that the election may have been rigged.
In chinese there is a saying, "when wind blows out of an empty cave, there may actually be reason behind it" (sorry for my limited english capability) and media have deliberately skipping those possible evidence.
Yes, they won't present to the viewers even those _possibilities_.
If you have nothing to cover up, why do the media skip everything that may lead to an open discussion of what may have happened with all those statistical-model-defying numbers and the opinion of / facts reported by the locals which have witness first hand what those voting centers have done?
It's just so much like the traditional communist take on voting.
Stalin once said, "I consider it completely unimportant who in the party will vote, or how; but what is extraordinarily important is this - who will count the votes, and how".
I am not personally against Biden or Trump - they both have policies appealing to different types of american. I am not an american and i think, ultimately, the americans are those who will feel deepest of the impact of their new president, and it is their choice. It is the open unfairness, or in some state, criminal unfairness, of the way votes are accepted and counted, and the mass media widespread attempt to mute the voices who questions the fairness of the election which makes this election so disputed.
Then the surveillance of the citizen's digital activities will be down to the OS level - combined with the password law enacted this year, the state can mandate what you can set as the root/user password of your linux box.
Next move will be a state-regulated bios and state-provided firmwire updates which, combined with state-issued linux distribution, will reflect the amount of control the state want at any given moment.
Why would anyone side with china on their digital policy which erode the citizen's privacy beyond the bedrock is beyond me.
From hong kong, where the freedom and rights is quickly aligning to the standard of the mainland.
The major reason why we arrive at the point we are at today is the western business just blindly offshore all the manufacturing needs to China for every cent that is saved for the profit of the western companies, and the bonus of the management of those companies. Design is based on expertise - you cannot design in a vacuum and expertise is gained through experience - and when the whole product development and delivery cycle, except the initial design step, is offshore to China how hard would it be for China to take over also the "initial" design steps? What would be left if this happens?
The western companies and, subsequently, the western culture are doomed for their pursuit (and their greediness) for the last cent in the drive of their profitability. Now reality is biting back. It will be a long long road ahead if anyone want to have a product development and delivery cycle (for any tangible thing) that completely bypass China - it is doable, it requires mega-cooperation of many organization and companies, and most important of all, requires the focus and persistence that is needed long enough for this to come to fruition and before the western people losing their memories of what got them here in the first place and returning to the status quo.
Sometimes I feel it all boils down to the greediness of the western companies and countries.
From a person in HK.
a little bit off-topic.
china's surveillance and control on citizen is much deeper than most westerners could imagine.
every thing you do in real life, keywords that you search on search engines, everything thing you said on anti-social networks, everything you buy online and everything else that are relatable to you contribute to a point-based system run by the state, which dictates what you can do, including if you are given suitable medical care and if you can travel abroad. if your points are too low then, sorry, you are in deep trouble.
every corner in big cities are covered with state-operated cameras - even bbc has a documentary that said once you are tagged as a criminal in the state-operated security system, it only take 7 minute for the police to find you out and arrest you.
enjoy the freedom that you are enjoying, and i really hope that your country won't become the next china.
if (updating the production/assembly line for current macbook pro for a better thermal/keyboard is so expensive &&
(
people complaining about failed keyboard is so few numerically so not justified to update ||
people complaining about thermal throttling is so few numerically so not justified to update ||
people buying macbook pro is so few numerically so not justified to update
)
){
keep churning out products from existing production/assembly line until it breaks or company faces class action;
}
(honestly, imo, even if they only fix the thermals and keyboard issues, it will already be a very nice machine)
I feel that this new language is still at its infancy - they are just designing the basic building blocks and it is quite far away from being usable in a problem-solving context.
They may need either:
(1) a killer application which have a common use-case (like rails for ruby), or
(2) a killer feature which have a common use-case (like channels for golang), or
(3) a killer reason to use it (like swift for iOS dev)
for it to pick up public interest.
Seems this language is not quite there yet, and i suspect that it will become a python wanna-be while light-years behind in terms of libraries and toolchains.
But reading half way through the paper, it do looks to have some novel approach to common pitfalls in programming and i feel, down the road, its syntax and concepts will sprinkle into some mainstream languages, like c# or java, to make dev life easier.
My level of productivity and number of people working around me is inversely related. Thanks boss for the flexibility, now I rent and work in my own tiny workshop not so far from my head office so that I can escape the productivity drop in that open plan office and get my job done promptly.
"Appealing to a broader church means making slight shifts – and one good example is the decision to offer multi-document ACID transactions, a feature that is generally available from today."
I don't know but, for us, the lack of multi document ACID transaction is exactly the reason why we only use it as logger instead of a wider data storage candidate. It is good to hear they put this into the engine but i just wonder if they implement this there will there be performance penalty? And does it really work as advertised? Honestly i think multi-document ACID transaction is no small undertaking and i only comfortable to use it in wider scope after, maybe, one or two major versions after it is rolled out.
sometimes i think... oracle needs to maintain a team of engineers for java so they should have the right to charge for it, and i don't think java, as a client technology, is popular by any metrics (other than android) so most of it is used within the server room, generating revenue, so i feel it is fair for oracle to charge for the business use of java (the java users are paying for the stack of middleware anyway... if ibm can charge for websphere, oracle should be able to charge for the runtime too).
and, as a developer using java for 15+ years, it may be a kind of... inertia that it seems not much new static-typed languages matches the consistency of language syntax and availability of good code-suggestion facilities in IDEs of java...
but i do think that for any free oracle technologies in use, it is rational to think that sooner or later oracle will charge for it and any non-free ones, expect to be charged increase significantly over its lifetime.
i do remember fondly the developer-friendly sun micro which created the whole java thing. i was a fan of their techs (solaris, sparc, java...) but oracle keep squeezing the last juice out of them and many of them are effectively eol'ed already. and a business as big as oracle have the revenue to make java free, if not just for a bit of positive publicity. maybe it is just the rule of business.
i feel programming in a stable (or, in some others' view, stagnant) programming language and api keeps us productive - at least i don't need to, like the javascript guys sitting around me, constantly learn new ways of solving the same type of problems when the api library makers re-invents itself (angularjs 1 -> 2 comes to mind). some people may argue that java is cumbersome (in terms of lines of code...) but it is static typed and the tools (like Intellij IDEA) already suggests or even fill in much of the codes for you (like generating getters/setters/anonymous classes) and, as a result, i feel i actually type less than when i code in other languages. the only thing i feel oracle should seriously work on is its GC engine, the performance of which i feel lags seriously behind other newer languages (like golang). for the java ee part, well, i think if any company can make an app server that is as user-friendly as glassfish server (the admin console is so easy to use, everything self-explaining) i think more people will flock into the java ee scene - personally, i feel the major roadblock to java app server is that most of them is hard to admin.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
