text
stringlengths
9
408
[113.12 --> 115.14] And they should have brought more people on board.
[115.42 --> 116.36] And they burned out.
[116.36 --> 123.62] You know, we got a really interesting insight into this manufacturing difficulties during our System76 factory tour last week,
[123.82 --> 131.88] where Jeremy, who is the guy that designed some of the firmware chips and some of the firmware itself for the launch keyboard,
[132.04 --> 139.34] was saying that they had to redesign on the fly two or three times some of the circuitry in that keyboard
[139.34 --> 142.36] because parts would go out of stock within hours.
[142.36 --> 149.88] And that he decided in the end to go ahead and order several, you know, hundred units of the parts they needed
[149.88 --> 155.20] before he'd actually finished the firmware so that it didn't go out of stock before he finished writing the code.
[155.38 --> 161.72] So if that's what System76 are up against, then they're probably a much larger player in this space than COBOL.
[162.60 --> 164.94] I can only imagine some of the issues that those guys are having.
[165.20 --> 169.36] It's a shame, though, because I think there's a lot of passion for an ARM-powered NAS,
[169.36 --> 172.88] specifically for those of us who are looking for low-power solutions.
[173.54 --> 177.18] We had some hopes that Helios was going to produce that for us.
[177.44 --> 180.96] Well, I think this speaks to some motivations that I've been feeling lately,
[181.12 --> 185.12] particularly after the last week getting the tour of the factory and seeing all the CNC machines
[185.12 --> 187.22] and all the powder coating stuff.
[187.40 --> 191.56] And, I mean, if ever you get the chance to be in Denver and do a tour of System76,
[192.44 --> 194.08] I'm not affiliated with them in any way.
[194.20 --> 195.86] You know, they're not a sponsor of the show or anything,
[196.36 --> 198.22] although they do sponsor other JB shows.
[198.22 --> 199.62] Just Coder, but yeah.
[200.12 --> 201.42] Oh, just Coder. That's right.
[201.88 --> 203.74] I mean, you know, maybe we should have them sponsor this show,
[203.82 --> 207.86] because those Thaleoses would make great NASes, so maybe they should be.
[208.70 --> 211.20] It was a truly inspirational experience for me.
[211.24 --> 216.40] So I'm sat here thinking, okay, COBOL, who was my hope, my light in this space,
[216.46 --> 218.20] has just bowed out.
[218.30 --> 223.16] Maybe Alex should start a company and build the ultimate NAS.
[223.36 --> 224.68] You know, maybe I should do that.
[224.68 --> 226.98] I think you've got a little bit of experience, Alex,
[227.08 --> 229.96] and some trials and tribulations that you've gone through already,
[230.08 --> 231.20] so maybe you're perfectly suited.
[231.46 --> 234.16] You could prototype 3D print the enclosure.
[234.52 --> 236.26] Well, you know, that was one of the things that struck me.
[236.32 --> 239.26] I know this is turning into a System76 factory tour recap,
[239.56 --> 242.70] but this is one of the things that struck me most during that tour, actually.
[242.78 --> 245.50] I asked them, you know, do you use 3D printing for your prototypes?
[245.94 --> 249.16] And they were like, well, no, it's just easier to do it in metal, isn't it?
[249.52 --> 250.36] And I'm like, is it?
[250.36 --> 251.96] Yeah, they used to.
[252.20 --> 253.48] Back in the day, they did.
[253.68 --> 256.40] I think one of the realities is that it's, with the equipment they have,
[256.44 --> 258.12] it's faster to make it out of metal.
[258.30 --> 260.88] It is really, you know, I think the reason why you keep coming back to it, Alex,
[260.94 --> 266.40] is because it is so hard to appreciate the amount of work that goes into manufacturing,
[266.68 --> 268.08] and that's an obvious thing to say.
[268.50 --> 273.80] But when you go to the factory, and you see every little thing they had to consider,
[273.80 --> 277.90] and when they're bringing things in as a block of aluminum,
[277.90 --> 280.76] and what's coming out on the other end is a keyboard or a case,
[281.44 --> 286.04] you really have no idea how intricate that process is until you see it.
[286.52 --> 290.54] And I think that's what is probably, I mean, that's what's resonated with me and stuck with me,
[290.56 --> 292.10] and I got to imagine probably for you, too.
[292.66 --> 296.16] Yeah, well, one example was Carl, who is the founder of System76.
[296.58 --> 300.40] I was just stood next to a shelf with loads of different parts in it,
[300.40 --> 304.26] and I picked up, like, an innocuous little nameplate, you know,
[304.28 --> 306.28] about the size of a dog tag or something like that,
[306.62 --> 310.22] and he jumped straight into a five-minute discussion telling me about all the different,
[310.38 --> 312.80] you know, reasons why they pick that particular size,
[313.26 --> 315.58] how it goes through their laser etching machine,
[315.74 --> 318.20] and how they're considering doing direct engraving instead,
[318.32 --> 319.12] and all that kind of stuff.
[319.14 --> 323.50] And I'm like, wow, if that's the amount of thought that goes into a nameplate,
[323.50 --> 324.92] or a serial number plate,
[325.80 --> 330.38] this is going to be a big mountain to solve if I want to try and do it myself.
[330.68 --> 332.44] That's what I take from it.
[332.48 --> 335.02] You take from it, hmm, maybe Alex could make a Nash.
[335.06 --> 335.72] That could be cool.
[335.82 --> 338.80] I take from it, oh, my God, I would have to build a factory,
[338.80 --> 340.70] and I would have to learn all this stuff.
[341.12 --> 342.70] Oh, well, that sounds fun, doesn't it?
[342.96 --> 344.18] Daunting, but fun.
[344.66 --> 346.88] Yeah, if I had another life,
[347.52 --> 350.60] if I had, like, you know, another hundred years,
[350.68 --> 352.74] I definitely would get into it for sure.
[353.20 --> 355.46] But you manufacture stuff every week as well,
[355.50 --> 357.04] and you ship podcasts every week.
[357.04 --> 361.46] Okay, you're not manufacturing aluminium into, you know, computer cases,
[362.00 --> 364.68] but you ship and manufacture stuff just the same.
[364.92 --> 366.02] That's true, and there have been many times
[366.02 --> 367.82] when we refer to our production pipeline
[367.82 --> 369.42] as kind of a manufacturing pipeline.
[369.66 --> 374.92] And if one component is delayed or, you know, gets out of order,
[375.34 --> 376.46] it does cause a jam up.
[376.54 --> 379.10] There is a lot of that kind of building tech around that.
[379.18 --> 381.20] Maybe that's why I don't have the appetite to do it again,
[381.20 --> 383.34] but I definitely would sit here and encourage you.
[383.64 --> 385.12] I'll be customer number one.
[386.92 --> 387.80] I'll be number two.
[388.28 --> 390.68] Well, there's a lot of roadblocks to overcome,
[391.30 --> 393.58] but if I could just bring it back to Cobol for a minute,
[393.84 --> 395.80] you know, one of the things I picked up on in my review
[395.80 --> 398.56] was that the product wasn't 100% finished,
[398.62 --> 400.32] and I've got to believe that with this news
[400.32 --> 403.56] that some of the issues that were present with that NAS now
[403.56 --> 405.10] will never be solved.
[405.24 --> 407.24] You know, things like the backplane breaking,
[407.24 --> 410.24] some of the plastic tabs, you know,
[410.28 --> 411.64] the power connector and data connector
[411.64 --> 413.82] on three of my hard drives, it turned out eventually.
[414.74 --> 417.06] Problems with the two and a half gigabit network adapter.