TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Technical Feasibility of Building Hitchhiker's Guide to the Galaxy

54 点作者 amund大约 11 年前

12 条评论

ctdonath大约 11 年前
All of Netflix, in HD, on 24TB.<p>The thought of putting most movies in my pocket for a few hundred bucks is ... stunning.<p>Netflix could solve the bandwidth problem by just <i>mailing the whole library</i> to customers. &quot;Never underestimate the bandwidth of a station wagon full of magnetic tapes.&quot;
评论 #7508493 未加载
评论 #7508492 未加载
评论 #7514637 未加载
评论 #7511197 未加载
dsr_大约 11 年前
Some information has value that remains relatively constant over time: historical records, literature. Some information has value that decays slowly: basic materials in the sciences, large bodies of well-established measurements.<p>Some information has value that decays fairly quickly: current scientific progress that invalidates older measurements or has better predictions than older theories, current events, prices of common items.<p>And some information has value that decays extremely quickly: weather, financial markets, prices for things that you want to buy or sell now, casual interactions with other people.<p>If you build a system that can connect fast enough to the Internet in terms of latency and throughput, you don&#x27;t need much local caching. That&#x27;s what we have in current smartphones.<p>Without a subEtha network, having a large local cache becomes increasingly important the farther away you are. But without a mission profile to plan for, estimating things in terms of current storage technology is as useless as coming up with a security plan without a threat model.
评论 #7512374 未加载
TeMPOraL大约 11 年前
If you plan on using it just on Earth, this becomes relevant: <a href="https://xkcd.com/548/" rel="nofollow">https:&#x2F;&#x2F;xkcd.com&#x2F;548&#x2F;</a> ;).
ajaimk大约 11 年前
The Netflix numbers are definitely wrong. Of the 10434 titles on there 3687 are TV Shows, or more specifically seasons of TV shows.<p>Assuming 2 hours&#x2F;movie and 15 hours&#x2F;tv show season, the number rolls around to 72,486 Hours of content. Let&#x27;s round that to 75,000 hours for convenience sake.<p>Assuming an average bitrate of 4Mbps, it comes to about 135 TB.<p>Which is why the Netflix Open Connect box works so well; Netflix can dump their entire catalog onto 2 of those boxes (well, it will take a few more since Netflix will be caching multiple versions of the files are different bitrates, but I wouldn&#x27;t be surprised if the entire catalog in all bitrates can fit into 1 rack)<p>Source for numbers: <a href="http://instantwatcher.com/" rel="nofollow">http:&#x2F;&#x2F;instantwatcher.com&#x2F;</a>
Theodores大约 11 年前
The creators of &#x27;Elite&#x27; had this problem way back in the 1980&#x27;s when they had to squeeze 8 galaxies complete with planet data into 32K of RAM. They solved the problem by &#x27;procedurally generating&#x27; all of the required data from a seed number. Therefore, it isn&#x27;t a question of cutting and pasting from Wikipedia etc., it is more a matter of getting the right &#x27;seed number&#x27; for earth, updating the procedure for generating content and that should be it.<p>Anyway, here is something cut &#x27;n&#x27; pasted from Wikipedia:<p><i>The Elite universe contains eight galaxies, each with 256 planets to explore. Due to the limited capabilities of 8-bit computers, these worlds are procedurally generated. A single seed number is run through a fixed algorithm the appropriate number of times and creates a sequence of numbers determining each planet&#x27;s complete composition (position in the galaxy, prices of commodities, and even name and local details— text strings are chosen numerically from a lookup table and assembled to produce unique descriptions for each planet). This means that no extra memory is needed to store the characteristics of each planet, yet each is unique and has fixed properties. Each galaxy is also procedurally generated from the first.<p>However, the use of procedural generation created a few problems. There are a number of poorly located systems that can be reached only by galactic hyperspace— these are more than 7 light years from their nearest neighbour, thus trapping the traveller. Braben and Bell also checked that none of the system names were profane - removing an entire galaxy after finding a planet named &quot;Arse&quot;.[9]</i><p>[9] Procedurally generated by unicorns.
评论 #7514660 未加载
protonfish大约 11 年前
So much of what makes the internet useful is the collaborative aspect. Having a copy of StackOverflow at a static point in time is better than nothing, but losing the ability to ask or answer new questions or stay up to date is a significant loss.<p>If you are a lone (or part of a small group) traveler, the thumb drive Hitchhiker&#x27;s Guide is probably as good as it gets. If we had a colony on Mars (or more distant) a better solution would be to have a planetary WAN (the MWW?) Central servers could update news, movies, apps and other timely but read-only content over a wide-bandwidth, high latency channel to the WWW (Renamed in 2250 to the Earth Wide Web.) Community sites (like HN and SO) would have to be redesigned for multiple non-interactive WANs. Perhaps it would default to your home WAN for voting and commenting, but you could click a tab to view a read-only, slightly older version of different planets&#x27; contributions.
analreceiver大约 11 年前
That&#x27;s all the information it would ideally have on earth, but then again, there are aproximately 8.8 BILLION habitable planets on our galaxy alone, and the more advanced ones oughta have a lot more information to store.<p>However, about our planet, the HHGTTG had only this to say: &quot;Mostly Harmless&quot;. It does have an entry for human beings though, so we actually don&#x27;t know how much information about us it contains.
评论 #7509885 未加载
DennisP大约 11 年前
A while back I had some thoughts about a user-generated travel guide, in the spirit of Hitchhikers, built like a wiki or the old IMDB. Has anyone heard of a project like that?<p>Maybe the problem is that it&#x27;d fill up with so much spam from local businesses that it&#x27;d be useless. Maybe a good reputation and rating system could make it work, if there were enough participation from neutral parties.
评论 #7508944 未加载
评论 #7508582 未加载
评论 #7510804 未加载
pyre大约 11 年前
1. The Wikipedia estimate probably only includes text. All of the media is distributed separately, and was ~200GB as of a few years ago. I can only imagine it&#x27;s grown since then. Some articles make more sense with images (e.g. a photo of an animal to go with the description).<p>2. Is the entirety of Twitter really useful as part of a Hitchhiker&#x27;s Guide to the Galaxy?
评论 #7508406 未加载
评论 #7508203 未加载
评论 #7507945 未加载
评论 #7507917 未加载
评论 #7510841 未加载
Raphael大约 11 年前
A Kindle with a copy of Wikipedia on it seems close enough to me.
dalek2point3大约 11 年前
but what about the legal side of things? last time I checked, Netflix wasn&#x27;t particularly excited about me trying to scrape all of their video data and making money off of it.
评论 #7510630 未加载
qwerta大约 11 年前
Perhaps we should start with interstellar propulsion system?