One question I've tried to answer is: has Iridium ever made enough money to even pay back the cost to put the satellites up. Using Google for all these rough numbers the first constellation cost $5 billion before Iridium (the first company) went bankrupt. For the second generation constellation launched between 2017 and 2019 it says $3 billion (for sats and launch). Compared to $400 million cumulative net income for Iridium (the second company) since bankruptcy restructuring ended in 2009. So as a non-investor (I only have boring index funds, no individual stocks) it seems like Iridium is a bad investment because it's a company that has spent 21+ years to turn $8 billion into $400 million (depending on when you want to start counting).
When Amazon bought Globalstar a couple months ago I had the same question and it's pretty much the same answer. For Globalstar there was basically 0 net income so the return on investment looked like it mostly came from spectrum gambling. Maybe that's the value for Iridium as well? Iridium does have some net income of around $100 million last year, but I don't know if RocketLab's vertical integration is going to be enough to flip the script. If RocketLab could have built and launched the Iridium Next constellation for $2 billion in 2017 would $100 million of net income 10 years later be a success?
That's how chapter 11 type bankruptcy works. The business continues to run but the debtors are now the owners. There's also chapter 7 where the business shuts down and stripped for parts to pay the debts.
Front light would make it the perfect device. I use it on the go everywhere now, to limit the distractions inherent with trying to read from my phone. But in bed having to pull out a book light or a headlight is an annoying step that risks disturbing my partner.
I ordered the Xteink X4 not knowing about CrossPoint and I was so disappointed with the included firmware:
- Almost no formatting on the ebooks you upload (no bold text, missing glyphs, no images/cover art)
- The book had to be rendered before being uploaded. I suspect they were uploading series of bitmaps.
- This meant if you wanted to change portrait/landscape or change font size/type you had to re-render the book
- Bitmap fonts were very ugly.
Overall it was barely workable. More of a proof of concept. The CrossPoint firmware on the other hand:
- Actually renders books as you would expect, in a pleasing manner, formatting and all
- Much more reader configuration on device: font settings, margin, spacing, alignment, progress display
- Update via USB or OTA
- Multiple ways to get your books: Connect to your Calibre library to push or pull books (thanks for introducing me to Calibre!), USB, Wireless file transfer
- Sync progress via KOReader
It's the best gift the community could give to the manufacturer. With only the default firmware my X4 would have been in the junk drawer within a week, but now I carry it every day, I've shown it to so many people. It's a marvel. With the news about them attempting to lock it down I can't recommend it anymore. Why would they do that?
> 3) What do you think ASML had to pay Lego to design and create this model? Or maybe Lego has a small division that does custom models like this? Do we know the employee cost for the Lego set?
I've done my own small scale version of this where I made models for internal distribution of the specialty equipment my employer uses. I doubt they paid Lego anything.
There is software to design your own lego set. Bricklink Studio is what I used. It's essentially Lego CAD software with a component library of Lego pieces. You can do high quality renderings, generate instruction sheets and BOMs.
Lego has a Pick-a-brick service where you can get new parts from a very limited selection at great expense. At third party marketplaces like Bricklink you can upload BOMs and they will assemble shopping carts from different seller's inventories of used pieces. Price and selection is better than Pick-a-brick but shipping / order fees / minimum lot size drive up the cost. I've tried many times but even the smallest 200 piece build ends up needing orders from 3 sellers across the world. There's always some part that was only ever sold in one rare set from 30 years ago and is unobtanium (the CAD program makes it easy to include a piece regardless of actual availability).
There are also businesses that will give you a turn key product that looks very retail-like with parts bagged by step, printed instructions, real Legos, box art, etc.
If you're willing to go with "lego compatible" third party bricks like GoBricks there are many sites that can source your entire build at once with new non-lego pieces. Part quality ranges from "good enough" to "indistinguishable" and the price and ease of ordering is loads better. You get a box of unsorted parts and spend lots of time grouping the pieces into kits.
Everyone is talking about the Playdate but I have a related Duke story about undergrad classes incorporating new hardware. My Digital Signal Processing course (ECE major) made a big deal about using these new things called iPods for class. Everyone got an iPod... for the semester. Even at Duke tuition prices you only got to borrow it. My recollection of the class work part was using a little piezo sensor that plugged into the microphone/headphone jack and recording your heart beat as a voice memo while doing a couple different activities. Maybe ten minutes for the semester. Then back at the computer doing a FFT to determine your heart rate. The lazy kids just got a copy of someone else's recording. This would have been 2004 or 2005. I think it was the third generation with clickwheel and monochrome screen.
They were about $3 billion in the hole when they went through bankruptcy in 2002 and the new owners bought it for $43 million (from Wikipedia). In 2025 they earned a return of $-8 million on that investment (plus any other money raised since then, such as $1 billion from Apple). So even the second incarnation doesn't seem to be a good business model even with free satellites.
The business model that works seems to be spectrum gambling. Do the minimum amount of satellite investment for decades until someone with a real business plan comes along and has to go through you to get it.
How did you make the "missing" corner? I'm assuming it wasn't chiseled out. In the last picture is the to-be-removed section made of plaster and then concrete poured around it?
Maybe average cost of next-size-up SSD price divided by a SWAG of a gaming PC lifetime? So if I had to buy a 2 TB NVMe stick instead of a 1 TB stick it's an extra $70 and I upgrade after 5 years that's only about $1 per TB-Month. I don't game I have no idea if those are good numbers.
The cheapest storage tier on s3 with instant retrieval is $.004 per GB-Month which implies AWS can still make money at $4 per TB-Month so $2.50 for consumer hardware sounds reasonable to me.
The incremental improvements to the engine thrust is par for the course. The exciting thing in this announcement is the new 9x4 configuration (9 and 4 engines in the first and second stages vs the current 7x2). They don't mention whether the tanks will get stretched to allow for more fuel, or if this just burns the fuel faster. Starship generations keep getting both more engines and longer.
When Amazon bought Globalstar a couple months ago I had the same question and it's pretty much the same answer. For Globalstar there was basically 0 net income so the return on investment looked like it mostly came from spectrum gambling. Maybe that's the value for Iridium as well? Iridium does have some net income of around $100 million last year, but I don't know if RocketLab's vertical integration is going to be enough to flip the script. If RocketLab could have built and launched the Iridium Next constellation for $2 billion in 2017 would $100 million of net income 10 years later be a success?