Stories
Slash Boxes
Comments

SoylentNews is people

What is the most overly over-hyped tech trend?

Displaying poll results.
Generative AI
  41% 38 votes
Quantum computing
  9% 9 votes
Blockchain, NFT, Cryptocurrency
  36% 34 votes
Edge computing
  1% 1 votes
Internet of Things
  4% 4 votes
6G
  2% 2 votes
I use the metaverse you insensitive clod
  3% 3 votes
Other (please specify in comments)
  1% 1 votes
92 total votes.
[ Voting Booth | Other Polls | Back Home ]
  • Don't complain about lack of options. You've got to pick a few when you do multiple choice. Those are the breaks.
  • Feel free to suggest poll ideas if you're feeling creative. I'd strongly suggest reading the past polls first.
  • This whole thing is wildly inaccurate. Rounding errors, ballot stuffers, dynamic IPs, firewalls. If you're using these numbers to do anything important, you're insane.
Display Options Threshold/Breakthrough Reply to Article Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 3, Interesting) by Freeman on Friday December 13, @06:35PM (14 children)

    by Freeman (732) on Friday December 13, @06:35PM (#1385365) Journal

    I voted for "Generative AI", if you had said "Current". It would have been a slam dunk in the Generative AI court. Though, I guess trend implies current.

    Still, cryptocurrency is definitely popular. Just look at the people getting scammed when *insert new meme coin* is introduced and about the only person who makes money is the person who created it.

    --
    Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
    • (Score: 5, Insightful) by mhajicek on Friday December 13, @08:09PM (11 children)

      by mhajicek (51) on Friday December 13, @08:09PM (#1385372)

      I voted cryptocurrency, because generative ai actually has legitimate uses, while cryptocurrency is 100% hype.

      --
      The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
      • (Score: 4, Interesting) by krishnoid on Saturday December 14, @02:40AM

        by krishnoid (1156) on Saturday December 14, @02:40AM (#1385392)

        Whether I believe in the hype or not, 2016 and beyond has taught me [wikipedia.org] if you can't beat 'em, see where people who support 'em are spending their money. So between cryptocurrency and generative AI, maybe nVidia is still a fair investment.

      • (Score: 3, Interesting) by krishnoid on Saturday December 14, @03:18AM (3 children)

        by krishnoid (1156) on Saturday December 14, @03:18AM (#1385393)

        Cryptocurrency sure, but blockchain [investopedia.com] definitely has its uses. Kind of like how BOINC was created (?) to run SETI@Home [wikipedia.org] but has since been used to support multiple other projects [berkeley.edu].

        • (Score: 5, Informative) by zocalo on Saturday December 14, @11:43AM (2 children)

          by zocalo (302) on Saturday December 14, @11:43AM (#1385405)
          I'll give you blockchain has some legit uses, but many of of those are already solved problems and adding a blockchain into the mix doesn't really add any value for anyone other than the suppliers of the blockchain engine and any hardware/cloud infrastructure to run it all on. The former is typically a consultancy[*], because the only kinds of organization that are large enough to actually think they could utilise blockchain for a private project tend to either have the expertise to do so in house, or use consultants and contractors for anything with a significant impact on business processes.

          [*] I work for a consultancy with a busines unit that offers blockchain solutions via a solution provider partnership. Almost all the blockchain projects I'm aware of are for the government bodies, military, or large industrial/service provider type organizations we generally work with looking to authenticate and track large asset/data/documentation sets, so maybe it's a different story in other sectors. While there have been a few "successes", most are are churning through a litany of problems of implementation, integration, adoption, etc., or have already been scrapped. I've yet to see a single case study of the so called successes that include any significant tangible benefits over the previous solution that are clearly quantifiable. The business unit and solution provider concerned make a profit on the job either way though, so it's all good, right?
          --
          UNIX? They're not even circumcised! Savages!
          • (Score: 1, Insightful) by Anonymous Coward on Saturday December 14, @11:51PM (1 child)

            by Anonymous Coward on Saturday December 14, @11:51PM (#1385462)

            As I said the last time this poll was up.

            At its base, the blockchain is an implementation of the basic technique to create a fault-tolerant, secure, distributed, tamper-resistant, open (anyone can read), and public (anyone can write) append-only database. Algorithms with some of those features have existed for quite some time with many different applications in use today. The big difference with blockchain is that it allows for both public and open access to the ledger because of a carefully chosen consensus algorithm. Things like blockchain have their uses. But finding one where being open is a benefit is already a small, add the additional factor of being public and the benefits are even rarer. This is because the open and public natures of blockchain adds quite a bit of theoretical complexity to the implementation that oftentimes outbalances the benefits it brings. Hence why most systems in use today give on at least one or the other or one of the other properties of a blockchain to balance out the various concerns.

            Unless you really need the database to have all 7 of those features, there isn't really a benefit over using one of the easier and cheaper alternatives.

            Or, as maxwell's demon suggested, maybe the author was a security researcher trying to generate a large body of data for a breakthrough in nested hash research.

            • (Score: 3, Interesting) by zocalo on Sunday December 15, @12:21PM

              by zocalo (302) on Sunday December 15, @12:21PM (#1385485)
              Yeah, totally agree, and despite us now being over sixteen years into this whole Bitcoin/blockchain thing, I still feel it's a solution in search of a problem. It's clever, certainly, but it comes with a significant overhead in hardware, software, and general complexity over existing systems. As you say; few of those problems are going to need all seven of those features in the solution (given that "public" would often translate to "a specified set of users"), and most, if not all, of the features they do need can be done with some process engineering, plus their existing relational DBs, and CMS (typically this will be Oracle/SQL Server, SharePoint, and PowerBI, but not always).

              AFAICT, the main problems our clients face isn't the size of the data sets, processes behind the data handling, and the like; it's ensuring data integrity and any required authorizations are present and correct within their existing DBs and document management systems. Typically, they've already ticked the boxes on fault-tolerant, secure (as far as practicable), distributed, and have access control more or less where it needs to be. That doesn't require a whole new system based on the blockchain; it just needs the existing processes be tweaked slightly and maybe the introduction of a few digital signatures here and there in conjunction with some more robust version control.
              --
              UNIX? They're not even circumcised! Savages!
      • (Score: 1, Interesting) by Anonymous Coward on Saturday December 14, @07:24AM (4 children)

        by Anonymous Coward on Saturday December 14, @07:24AM (#1385399)

        What is the legitimate use of mass generated sentences with no semantic content, at best filled with false information plausibly presented? Lorum ipsum is a solved problem, and canonical trolls abound on the internet.

        Consider Chomsky's dragon: "Colourless green ideas sleep furiously." Grammatically perfect. Zero meaning (or truth).

        Reminder that LLMs are a scam, and you are not in on the con. But it serves the conmen if you think you are.

        • (Score: 2) by mhajicek on Saturday December 14, @06:04PM (2 children)

          by mhajicek (51) on Saturday December 14, @06:04PM (#1385430)

          One anecdote: I run a CNC machine shop. I've been able to successfully use Claude as a sounding board for potential solutions to machining difficulties. It seems to understand things like how the different cutter coatings interact with different workpiece materials, and many other relevant details.

          --
          The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
          • (Score: 3, Touché) by Anonymous Coward on Sunday December 15, @08:15AM (1 child)

            by Anonymous Coward on Sunday December 15, @08:15AM (#1385483)

            It seems to understand. The problem is that it provably doesn't. Lucky for you, the people it is copying do and so the text it generates embeds that understanding. But then you ask it a question you don't know the answer to beforehand. How are you to tell apart instances when it copies the speech patterns of people who know and when it copies the speech patterns of those that either don't, were talking about something completely different with similar words to your question, or just plain made a mistake. That is the real problem because without some sort of knowledge-based check on the answers, you are just hoping the person in the Chinese Room or the computer-generated instructions he consults therein don't error.

            • (Score: 3, Touché) by mhajicek on Sunday December 15, @05:14PM

              by mhajicek (51) on Sunday December 15, @05:14PM (#1385506)

              I can tell if it makes a mistake because I'm asking it about a subject I'm an expert in.

              --
              The spacelike surfaces of time foliations can have a cusp at the surface of discontinuity. - P. Hajicek
        • (Score: 2) by Freeman on Monday December 16, @04:08PM

          by Freeman (732) on Monday December 16, @04:08PM (#1385612) Journal

          Generating good enough e-mail filler, without needing to come up with it yourself. Also, it's pretty good at coming up with basic programming solutions. Certainly faster than me searching for them. Sure, it's not going to do all of the work for you. However using it as a time saver certainly works!

          --
          Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
      • (Score: 4, Interesting) by zocalo on Saturday December 14, @11:23AM

        by zocalo (302) on Saturday December 14, @11:23AM (#1385402)
        Same here. The real problem with GAI isn't that it hallucinates, it's that users of GAI just shovel the slop it produces out without vetting it first, which then feeds back into future GAI content because the various engines eventually end up getting trained on the slop, and it's all downhill from there. Put it into a closed system and some of the results can be amazing, especially around image enhacement and content generation, analysis of large raw data sets, and so on.

        Frankly, I'm surprised OpenAI et al haven't already tried to break the feedback loop by making sure AI generated content is tagged somehow, either through support of legal efforts to make sure AI-generated content is visibly tagged as such, and potentially through some form of steganography or markup embedded within the generated content as well (admittedly hard with sort snippets of text). Some of the stuff being put out now is clearly garbage to anyone with a clue about the subject matter (which, unfortuntely, still leaves a awful lot of consumers who can't tell), but my feeling is that it's not going to take too many more training iterations before the mountain of AI generated hallucinations and other slop overwhelms the genunine content and the datasets become useless and impossible to sanitise.
        --
        UNIX? They're not even circumcised! Savages!
    • (Score: 4, Insightful) by khallow on Tuesday December 17, @02:39AM (1 child)

      by khallow (3766) Subscriber Badge on Tuesday December 17, @02:39AM (#1385663) Journal
      I voted Internet of Things because everything else on that list has little to no value, but IoT is about exploiting its customers - it's actively harmful. Let's spend a lot of money installing spies in our own homes.
      • (Score: 3, Informative) by zocalo on Tuesday December 17, @02:32PM

        by zocalo (302) on Tuesday December 17, @02:32PM (#1385697)
        IoT goes beyond the insecure electronic shovelware that grabs all your data and, once they've done that, any support vanishes because a supplier goes bankrupt or simply decides to pull the plug on the cloud servers required for it to work because they have a newer version (YMMV on whether it's also better). It's proving massively successful at replacing large scale telemetry networks; think swapping several hundred dollars worth of PLCs and routers/modems with something essentially built on a dev board like a Raspberry Pi at hundreds, or even thousands, of monitoring points. It's also critical to implementing at-scale (office block/campus, rather than home) smart energy systems, e.g. turning suitable devices on/off, or putting them into a reduced power state, that you're not drawing - and paying for - any more power from the grid than necessary when you're not generating enough locally.

        I've worked on examples of both of the above, and there are very real cost savings and efficiency benefits to be had - user facing views of the data in realtime too, if you want them (gamification of power saving can make a huge difference in how much energy, and therefore money, you can save).
        --
        UNIX? They're not even circumcised! Savages!
  • (Score: 5, Insightful) by Hartree on Tuesday December 17, @02:38AM (1 child)

    by Hartree (195) on Tuesday December 17, @02:38AM (#1385662)

    All of the above.

  • (Score: 3, Informative) by drussell on Tuesday December 17, @09:32AM (1 child)

    by drussell (2678) on Tuesday December 17, @09:32AM (#1385679) Journal

    Are we somehow expecting significantly different answers than the last time this same poll was run, all the way back in June?

    -1 Redundant Poll

    ;-)

    • (Score: 2) by janrinok on Tuesday December 17, @10:03AM

      by janrinok (52) Subscriber Badge on Tuesday December 17, @10:03AM (#1385682) Journal

      Please submit you suggestions, with suggested answers, and we will use it if at all possible.

      --
      I am not interested in knowing who people are or where they live. My interest starts and stops at our servers.
(1)